Author: "Scientific Data Management (ZENITH)" - Searchworks@Jio Institute Digital Library Search Results

Your search keyword '"Scientific Data Management (ZENITH)"' showing total 615 results

Start Over Author "Scientific Data Management (ZENITH)"

615 results on '"Scientific Data Management (ZENITH)"'

1. Distributed KNN-graph approximation via hashing

Author: ZENITH: Scientific Data Management (ZENITH) ; INRIA - Département Informatique (INFO/LIRMM) ; Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM) ; Université de Montpellier (UM) - CNRS - Université de Montpellier (UM) - CNRS - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM) ; Université de Montpellier (UM) - CNRS - Université de Montpellier (UM) - CNRS, INRIA Saclay - Ile de France (INRIA Saclay - Ile de France) ; INRIA, European Project : 288720, ICT, FP7-ICT-2011-7, GLOCAL - ENLARGED EU(2011), Trad, Riadh, Joly, Alexis, Nozha, Boujemaa, ZENITH: Scientific Data Management (ZENITH) ; INRIA - Département Informatique (INFO/LIRMM) ; Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM) ; Université de Montpellier (UM) - CNRS - Université de Montpellier (UM) - CNRS - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM) ; Université de Montpellier (UM) - CNRS - Université de Montpellier (UM) - CNRS, INRIA Saclay - Ile de France (INRIA Saclay - Ile de France) ; INRIA, European Project : 288720, ICT, FP7-ICT-2011-7, GLOCAL - ENLARGED EU(2011), Trad, Riadh, Joly, Alexis, and Nozha, Boujemaa
Abstract: International audience, Efficiently constructing the K-Nearest Neighbor Graph (K-NNG) of large and high dimensional datasets is crucial for many applications with feature-rich objects, such as images or other multimedia content. In this paper we investigate the use of high dimensional hashing methods for efficiently approximating the K-NNG, notably in distributed environments. We first discuss the importance of balancing issues on the performance of such approaches and show why the baseline approach using Locality Sensitive Hashing does not perform well. Our new KNN-join method is based on RMMH, a recently introduced hash function family based on randomly trained classifiers. We show that the resulting hash tables are much more balanced and that the number of resulting collisions can be greatly reduced without degrading quality. We further improve the load balancing of our distributed approach by designing a parallelized local join algorithm, implemented within the MapReduce framework.

2. High-Dimensional Private Empirical Risk Minimization by Greedy Coordinate Descent

Author: Mangold, Paul, Bellet, Aurélien, Salmon, Joseph, Tommasi, Marc, Machine Learning in Information Networks (MAGNET), Inria Lille - Nord Europe, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), Institut Montpelliérain Alexander Grothendieck (IMAG), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Institut Universitaire de France (IUF), Ministère de l'Education nationale, de l’Enseignement supérieur et de la Recherche (M.E.N.E.S.R.), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), ANR-20-CE23-0015,PRIDE,Apprentissage automatique décentralisé et préservant la vie privée(2020), ANR-20-CHIA-0001,CAMELOT,Apprentissage automatique et optimisation coopératifs.(2020), ANR-22-PECY-0002,iPoP,interdisciplinary Project on Privacy(2022), Machine Learning in Information Networks [MAGNET], Institut Montpelliérain Alexander Grothendieck [IMAG], Institut Universitaire de France [IUF], and Scientific Data Management [ZENITH]
Subjects: FOS: Computer and information sciences, Computer Science - Machine Learning, [INFO.INFO-CR]Computer Science [cs]/Cryptography and Security [cs.CR], Computer Science - Cryptography and Security, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], Statistics - Machine Learning, Machine Learning (stat.ML), Cryptography and Security (cs.CR), Machine Learning (cs.LG)
Abstract: In this paper, we study differentially private empirical risk minimization (DP-ERM). It has been shown that the (worst-case) utility of DP-ERM reduces as the dimension increases. This is a major obstacle to privately learning large machine learning models. In high dimension, it is common for some model's parameters to carry more information than others. To exploit this, we propose a differentially private greedy coordinate descent (DP-GCD) algorithm. At each iteration, DP-GCD privately performs a coordinate-wise gradient step along the gradients' (approximately) greatest entry. We show theoretically that DP-GCD can improve utility by exploiting structural properties of the problem's solution (such as sparsity or quasi-sparsity), with very fast progress in early iterations. We then illustrate this numerically, both on synthetic and real datasets. Finally, we describe promising directions for future work.
Published: 2023

3. Deep neural network based multichannel audio source separation

Author: Emmanuel Vincent, Aditya Arie Nugraha, Antoine Liutkus, Speech Modeling for Facilitating Oral-Based Communication ( MULTISPEECH ), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique ( Inria ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -Department of Natural Language Processing & Knowledge Discovery ( LORIA - NLPKD ), Laboratoire Lorrain de Recherche en Informatique et ses Applications ( LORIA ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Université de Lorraine ( UL ) -Centre National de la Recherche Scientifique ( CNRS ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -Université de Lorraine ( UL ) -Centre National de la Recherche Scientifique ( CNRS ) -Laboratoire Lorrain de Recherche en Informatique et ses Applications ( LORIA ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Université de Lorraine ( UL ) -Centre National de la Recherche Scientifique ( CNRS ) -Université de Lorraine ( UL ) -Centre National de la Recherche Scientifique ( CNRS ), Scientific Data Management ( ZENITH ), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier ( LIRMM ), Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ) -Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ) -Inria Sophia Antipolis - Méditerranée ( CRISAM ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Institut National de Recherche en Informatique et en Automatique ( Inria ), Speech Modeling for Facilitating Oral-Based Communication (MULTISPEECH), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL), Scientific Data Management (ZENITH), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Grid'5000, ANR-15-CE38-0003,KAMoulox,Démixage en ligne de larges archives sonores(2015), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), and Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Inria Sophia Antipolis - Méditerranée (CRISAM)
Subjects: [ INFO.INFO-TS ] Computer Science [cs]/Signal and Image Processing, Artificial neural network, Computer science, Wiener filter, Computer Science::Neural and Evolutionary Computation, 020206 networking & telecommunications, 02 engineering and technology, Non-negative matrix factorization, Speech enhancement, 030507 speech-language pathology & audiology, 03 medical and health sciences, symbols.namesake, Task (computing), [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, 0202 electrical engineering, electronic engineering, information engineering, Source separation, symbols, 0305 other medical science, Spatial analysis, Algorithm, Gaussian network model
Abstract: International audience; This chapter presents a multichannel audio source separation framework where deep neural networks (DNNs) are used to model the source spectra and combined with the classical multichannel Gaussian model to exploit the spatial information. The parameters are estimated in an iterative expectation-maximization (EM) fashion and used to derive a multichannel Wiener filter. Different design choices and their impact on the performance are discussed. They include the cost functions for DNN training, the number of parameter updates, the use of multiple DNNs, and the use of weighted parameter updates. Finally, we present its application to a speech enhancement task and a music separation task. The experimental results show the benefit of the multichannel DNN-based approach over a single-channel DNN-based approach and the multichannel nonnegative matrix factorization based iterative EM framework.
Published: 2018

4. The role of hydraulics FSPMs in the context of root breeding : a case study on Pearl Millet

Author: Ndour, Adama, Christophe Pradal, Vadez, Vincent, Passot, Sixtine, Guédon, Yann, Laplaze, Laurent, Lucas, Mikaël, LMI Adaptation des Plantes et microorganismes associés aux Stress Environnementaux [Dakar] (LAPSE), Institut de Recherche pour le Développement (IRD), Amélioration génétique et adaptation des plantes méditerranéennes et tropicales (UMR AGAP), Institut National de la Recherche Agronomique (INRA)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro), Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Scientific Data Management (ZENITH), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), International Crops Research Institute for the Semi-Arid Tropics [Inde] (ICRISAT), Consultative Group on International Agricultural Research [CGIAR] (CGIAR), Modeling plant morphogenesis at different scales, from genes to phenotype (VIRTUAL PLANTS), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de la Recherche Agronomique (INRA)-Amélioration génétique et adaptation des plantes méditerranéennes et tropicales (UMR AGAP), Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de la Recherche Agronomique (INRA)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de la Recherche Agronomique (INRA)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro), Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Inria Sophia Antipolis - Méditerranée (CRISAM), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de la Recherche Agronomique (INRA)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de la Recherche Agronomique (INRA)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro), LMI Adaptation des Plantes et microorganismes associés aux Stress Environnementaux [Dakar] ( LAPSE ), Institut de Recherche pour le Développement ( IRD ), Amélioration génétique et adaptation des plantes méditerranéennes et tropicales ( UMR AGAP ), Institut national de la recherche agronomique [Montpellier] ( INRA Montpellier ) -Centre international d'études supérieures en sciences agronomiques ( Montpellier SupAgro ) -Centre de Coopération Internationale en Recherche Agronomique pour le Développement ( CIRAD ) -Institut national d’études supérieures agronomiques de Montpellier ( Montpellier SupAgro ), Scientific Data Management ( ZENITH ), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier ( LIRMM ), Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ) -Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ) -Inria Sophia Antipolis - Méditerranée ( CRISAM ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Institut National de Recherche en Informatique et en Automatique ( Inria ), International Crops Research Institute for the Semi-Arid Tropics [Patancheru] ( ICRISAT ), Modeling plant morphogenesis at different scales, from genes to phenotype ( VIRTUAL PLANTS ), Inria Sophia Antipolis - Méditerranée ( CRISAM ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -Institut National de la Recherche Agronomique ( INRA ) -Centre de coopération internationale en recherche agronomique pour le développement [CIRAD] : UMR51, Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro)-Institut National de la Recherche Agronomique (INRA)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro), International Crops Research Institute for the Semi-Arid Tropics [Patancheru] (ICRISAT), and Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro)-Institut National de la Recherche Agronomique (INRA)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro)-Institut National de la Recherche Agronomique (INRA)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Inria Sophia Antipolis - Méditerranée (CRISAM)
Subjects: [ INFO.INFO-MO ] Computer Science [cs]/Modeling and Simulation, [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation
Abstract: International audience; Developing a sustainable agricultural model is one of the great challenges of the coming years. The agricultural practices inherited from the Green Revolution of the 1960s show their limits today, and new paradigms need to be explored in the context of counter rising issues such as the multiplication of climate-change related drought episodes. Two such new paradigms are the use of functional-structural plant models to complement and rationalize breeding approaches and a renewed focus on root systems as untapped sources of plant improvement. Since the late 1980s, numerous functional and structural models of root systems were developed and used to investigate the properties of root systems in soil or lab-conditions. In this talk, we present a review on the use of multiscale functional-structural hydraulic root models in the context of drought tolerance breeding. We discuss how root models predictions can be linked to breeding studies to improve plant resistance to drought and how they can be validated to demonstrate models reliability and use. To illustrate this topic, we present a new structural model of pearl millet root system growth dynamics, combining stochastic and data-driven modules. The model is capable of simulating the development of observed root phenotypic variability of two millet genotypes chosen for their contrasted root traits. Model description, principle, assumptions, formalism and simulations will be presented during the talk.
Published: 2018

5. Rank aggregation with ties

Author: Sylvie Hamel, Bryan Brancotte, Guillaume Blin, Sarah Cohen-Boulakia, Alain Denise, Bo Yang, Algorithms and Models for Integrative Biology (AMIB ), Laboratoire d'informatique de l'École polytechnique [Palaiseau] (LIX), École polytechnique (X)-Centre National de la Recherche Scientifique (CNRS)-École polytechnique (X)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), State Key Laboratory of Virology, Wuhan University [China], Laboratoire Bordelais de Recherche en Informatique (LaBRI), Université de Bordeaux (UB)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Centre National de la Recherche Scientifique (CNRS), Département d'Informatique et de Recherche Opérationnelle [Montreal] (DIRO), Université de Montréal (UdeM), Université de Bordeaux (UB)-Centre National de la Recherche Scientifique (CNRS)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB), Institut de Biologie Computationnelle (IBC), Université de Montpellier (UM)-Institut National de la Recherche Agronomique (INRA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS), Scientific Data Management (ZENITH), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Inria Sophia Antipolis - Méditerranée (CRISAM), Institut de Biologie Intégrative de la Cellule (I2BC), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), PEPS RankaBio, Centre National de la Recherche Scientifique (CNRS)-École polytechnique (X)-Centre National de la Recherche Scientifique (CNRS)-École polytechnique (X)-Laboratoire de Recherche en Informatique (LRI), Centre National de la Recherche Scientifique (CNRS)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Université Sciences et Technologies - Bordeaux 1-Université Bordeaux Segalen - Bordeaux 2, Institut National de la Recherche Agronomique (INRA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS), Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Inria Sophia Antipolis - Méditerranée (CRISAM), Université Paris-Sud - Paris 11 (UP11)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), Université de Montréal [Montréal], Université Paris-Sud - Paris 11 (UP11)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Centre National de la Recherche Scientifique (CNRS)-Université Paris-Saclay, Laboratoire de Recherche en Informatique ( LRI ), Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ), Laboratoire Bordelais de Recherche en Informatique ( LaBRI ), Centre National de la Recherche Scientifique ( CNRS ) -École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Université Sciences et Technologies - Bordeaux 1-Université Bordeaux Segalen - Bordeaux 2, Institut de Biologie Computationnelle ( IBC ), Centre de Coopération Internationale en Recherche Agronomique pour le Développement ( CIRAD ) -Institut National de la Recherche Agronomique ( INRA ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ), Scientific Data Management ( ZENITH ), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier ( LIRMM ), Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ) -Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ) -Inria Sophia Antipolis - Méditerranée ( CRISAM ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Institut National de Recherche en Informatique et en Automatique ( Inria ), Institut de Biologie Intégrative de la Cellule ( I2BC ), Université Paris-Saclay-Centre National de la Recherche Scientifique ( CNRS ) -Commissariat à l'énergie atomique et aux énergies alternatives ( CEA ) -Université Paris-Sud - Paris 11 ( UP11 ), Département d'Informatique et de Recherche Opérationnelle [Montreal] ( DIRO ), and Université de Montréal
Subjects: [ INFO.INFO-IR ] Computer Science [cs]/Information Retrieval [cs.IR], [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], Computer science, [INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS], Rank (computer programming), General Engineering, Scale (descriptive set theory), 0102 computer and information sciences, 02 engineering and technology, computer.software_genre, 01 natural sciences, [ INFO.INFO-DB ] Computer Science [cs]/Databases [cs.DB], Ranking, 010201 computation theory & mathematics, Position (vector), [ INFO.INFO-BI ] Computer Science [cs]/Bioinformatics [q-bio.QM], [INFO.INFO-IR]Computer Science [cs]/Information Retrieval [cs.IR], 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Data mining, [INFO.INFO-BI]Computer Science [cs]/Bioinformatics [q-bio.QM], computer, [ INFO.INFO-DS ] Computer Science [cs]/Data Structures and Algorithms [cs.DS], ComputingMilieux_MISCELLANEOUS
Abstract: The problem of aggregating multiple rankings into one consensus ranking is an active research topic especially in the database community. Various studies have implemented methods for rank aggregation and may have come up with contradicting conclusions upon which algorithms work best. Comparing such results is cumbersome, as the original studies mixed different approaches and used very different evaluation datasets and metrics. Additionally, in real applications, the rankings to be aggregated may not be permutations where elements are strictly ordered, but they may have ties where some elements are placed at the same position. However, most of the studies have not considered ties. This paper introduces the first large scale study of algorithms for rank aggregation with ties. More precisely, (i) we review rank aggregation algorithms and determine whether or not they can handle ties; (ii) we propose the first implementation to compute the exact solution of the Rank Aggregation with ties problem; (iii) we evaluate algorithms for rank aggregation with ties on a very large panel of both real and carefully generated synthetic datasets; (iv) we provide guidance on the algorithms to be favored depending on dataset features.
Published: 2015

6. DPiSAX: Massively Distributed Partitioned iSAX

Author: Florent Masseglia, Reza Akbarinia, Djamel Edine Yagoubi, Themis Palpanas, Scientific Data Management (ZENITH), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Institut de Biologie Computationnelle (IBC), Institut National de la Recherche Agronomique (INRA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS), Université Paris Descartes - Paris 5 (UPD5), Université de Montpellier (UM)-Institut National de la Recherche Agronomique (INRA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Inria Sophia Antipolis - Méditerranée (CRISAM), Scientific Data Management ( ZENITH ), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier ( LIRMM ), Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ) -Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ) -Inria Sophia Antipolis - Méditerranée ( CRISAM ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Institut National de Recherche en Informatique et en Automatique ( Inria ), Institut de Biologie Computationnelle ( IBC ), Centre de Coopération Internationale en Recherche Agronomique pour le Développement ( CIRAD ) -Institut National de la Recherche Agronomique ( INRA ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ), and Université Paris Descartes - Paris 5 ( UPD5 )
Subjects: High interest, [ INFO.INFO-NA ] Computer Science [cs]/Numerical Analysis [cs.NA], Computer science, Search engine indexing, Parallel algorithm, 02 engineering and technology, Load balancing (computing), [INFO.INFO-NA]Computer Science [cs]/Numerical Analysis [cs.NA], computer.software_genre, Temporal database, Database index, [ INFO.INFO-DC ] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], 020204 information systems, 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Data mining, [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], computer
Abstract: International audience; Indexing is crucial for many data mining tasks that rely on efficient and effective similarity query processing. Consequently, indexing large volumes of time series, along with high performance similarity query processing, have became topics of high interest. For many applications across diverse domains though, the amount of data to be processed might be intractable for a single machine, making existing centralized indexing solutions inefficient. We propose a parallel indexing solution that gracefully scales to billions of time series, and a parallel query processing strategy that, given a batch of queries, efficiently exploits the index. Our experiments, on both synthetic and real world data, illustrate that our index creation algorithm works on 1 billion time series in less than 2 hours , while the state of the art centralized algorithms need more than 5 days. Also, our distributed querying algorithm is able to efficiently process millions of queries over collections of billions of time series, thanks to an effective load balancing mechanism.
Published: 2017

7. The GeoLifeCLEF 2023 Dataset to evaluate plant species distribution models at high spatial resolution across Europe

Author: Botella, Christophe, Benjamin, Deneu, Diego Gonzalez, Marcos, Maximilien, Servajean, Théo, Larcher, Joaquim, Estopinan, César, Leblanc, Bonnet, Pierre, Joly, Alexis, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), ADVanced Analytics for data SciencE (ADVANSE), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Université Paul-Valéry - Montpellier 3 (UPVM), Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), European Project: 101060693,GUARDEN, and European Project: 101060639,MAMBO
Subjects: Europe, [SDV.EE]Life Sciences [q-bio]/Ecology, environment, Presence-only, [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], SDM, Deep SDM, prediction, [SDV.BID]Life Sciences [q-bio]/Biodiversity, Remote sensing, Presence-absence, Species distribution, species assemblages, biodiversity
Abstract: The full dataset is freely available at the link below (perennial repository) for academic use or other non-commercial use: https://lab.plantnet.org/seafile/d/936fe4298a5a4f4c8dbd/; The difficulty to measure or predict species community composition at fine spatio-temporal resolution and over large spatial scales severely hampers our ability to understand species assemblages and take appropriate conservation measures. Despite the progress in species distribution modeling (SDM) over the past decades, SDM are just beginning to integrate high resolution remote sensing data and their predictions are still entailed by the many biases and heterogeneous quality of the available biodiversity observations, most often opportunistic presence only data. We designed a European scale dataset covering around 10K plant species to calibrate and evaluate SDM predictions of species composition in space and time at high spatial resolution (~10m), and study their spatial transferability. For model training, we extracted and harmonized 5 million heterogeneous presence-only observations from selected GBIF datasets and 5 thousand exhaustive presence-absence surveys both sampled during the 2017-2021. We associated them to various environmental rasters classically used in SDMs, as well as to 10x10m resolution RGB and Near-Infra-Red satellite images and 20 years-time series of climatic variables and satellite point values. The evaluation dataset is based on 20K standardized presence-absence surveys separated from the training set with a spatial block hold out procedure. The GeoLifeCLEF 2023 dataset is open access and the first benchmark for researchers aiming to improve the prediction of plant species composition at a very fine spatial grain and at continental scale. It is also a space to explore new ways of combining massive and diverse species observations with environmental information at various scales. Innovative AI-based approaches, in particular, should be among the most interesting methods to experiment with on the GeoLifeCLEF 2023 dataset.
Published: 2023

8. Spatio-temporal analysis of strawberry architecture: insights into the control of branching and inflorescence complexity

Author: Marc Labadie, Karine Guy, Marie-Noëlle Demené, Yves Caraglio, Gaetan Heidsieck, Amelia Gaston, Christophe Rothan, Yann Guédon, Christophe Pradal, Béatrice Denoyes, Amélioration génétique et adaptation des plantes méditerranéennes et tropicales (UMR AGAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro Montpellier, Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Université de Montpellier (UM), Biologie du fruit et pathologie (BFP), Université de Bordeaux (UB)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), Invenio, Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)
Subjects: plant architecture, Physiology, OpenAlea, [SDV]Life Sciences [q-bio], branching, multiscale tree graph, [SDV.BV]Life Sciences [q-bio]/Vegetal Biology, Plant Science, Fragaria × ananassa, hidden hybrid Markov/semi-Markov chain
Abstract: Plant architecture plays a major role in flowering and therefore in crop yield. Attempts to visualize and analyse strawberry plant architecture have been few to date. Here, we developed open-source software combining two- and three-dimensional representations of plant development over time along with statistical methods to explore the variability in spatio-temporal development of plant architecture in cultivated strawberry. We applied this software to six seasonal strawberry varieties whose plants were exhaustively described monthly at the node scale. Results showed that the architectural pattern of the strawberry plant is characterized by a decrease of the module complexity between the zeroth-order module (primary crown) and higher-order modules (lateral branch crowns and extension crowns). Furthermore, for each variety, we could identify traits with a central role in determining yield, such as date of appearance and number of branches. By modeling the spatial organization of axillary meristem fate on the zeroth-order module using a hidden hybrid Markov/semi-Markov mathematical model, we further identified three zones with different probabilities of production of branch crowns, dormant buds, or stolons. This open-source software will be of value to the scientific community and breeders in studying the influence of environmental and genetic cues on strawberry architecture and yield.
Published: 2023

9. Methodologies for Reproducible Analysis of Workflows on the Edge-to-Cloud Continuum

Author: Rosendo, Daniel, Scalable Storage for Clouds and Beyond (KerData), Inria Rennes – Bretagne Atlantique, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-SYSTÈMES LARGE ÉCHELLE (IRISA-D1), Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-IMT Atlantique (IMT Atlantique), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT)-Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-IMT Atlantique (IMT Atlantique), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), HPC-BigData Inria Challenge, INSA RENNES, Gabriel Antoniu, Alexandru Costan, Patrick Valduriez, and ANR-15-CE25-0003,OverFlow,Workflow Data Management as a Service pour des Applications Multi-Site(2015)
Subjects: Méthodologie, [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], Edge-to-Cloud Continuum, Methodology, Reproducibility, Workflows, Continuum Edge-to-Cloud, [INFO.INFO-NI]Computer Science [cs]/Networking and Internet Architecture [cs.NI], [INFO.INFO-PF]Computer Science [cs]/Performance [cs.PF], [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Edge, E2Clab, [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], Cloud, Reproductibilité
Abstract: Distributed infrastructures for computation and analytics are now evolving towards an interconnected ecosystem allowing complex applications to be executed on the Edge-to-Cloud continuum. Understanding and optimizing end-to-end performance in such a complex continuum is challenging. One crucial challenge is accurately reproducing relevant behaviors of a given application workflow and representative settings of the physical infrastructure underlying this complex continuum. This thesis is a conceptual and practical contribution to the Edge-to-Cloud continuum, proposing and applying methodologies in novel environments. Our methodologies aim at overcoming the complexity of understanding and optimizing Edge-to-Cloud workflows. As such, they enable reproducible experiment design, application optimization, efficient workflow provenance capture, and cost-effective experiment reproducibility. We validated our proposal by first developing the E2Clab framework that supports the complete analysis cycle of an application on the Edge-to-Cloud Continuum and then using E2Clab to optimize Pl@ntNet, a global plant identification application. Large-scale experimental validation on Grid'5000 shows that our methodology has proven helpful for understanding and improving the performance of Pl@ntNet.; Les infrastructures distribuées pour le calcul et l'analyse évoluent désormais vers un écosystème interconnecté permettant l'exécution d'applications complexes dans le continuum Edge-to-Cloud. Comprendre et optimiser les performances de bout en bout dans ce contexte est un défi majeur. Un besoin crucial consiste à reproduire avec précision les comportements pertinents des workflows et les paramètres représentatifs de l'infrastructure physique sous-jascente.Cette thèse est une contribution conceptuelle et pratique au continuum Edge-to-Cloud, proposant des méthodologies et les appliquant dans des environnements novateurs. Nos méthodologies ont pour but de s'affranchir de la complexité de la compréhension et de l'optimisation des workflows dans le continuum Edge-to-Cloud. Ainsi, elles permettent la conception d'expériences reproductibles, l'optimisation des applications, la capture efficace des données de provenance des exécutions de workflows, et la reproductibilité des expériences. Nous avons validé notre proposition avec, d'abord, le développement du framework E2Clab qui supporte le cycle complet d’analyse d’une application dans le continuum Edge-to-Cloud, et ensuite, l'utilisation de E2Clab pour l'optimisation Pl@ntNet, une application mondiale d'identification des plantes. La validation expérimentale à grande échelle sur Grid'5000 montre que notre méthodologie s'est avérée utile pour comprendre et améliorer les performances de Pl@ntNet.
Published: 2023

10. MoveFormer: a Transformer-based model for step-selection animal movement modelling

Author: Ondřej Cífka, Simon Chamaillé-Jammes, Antoine Liutkus, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Université de Montpellier (UM), Centre d’Ecologie Fonctionnelle et Evolutive (CEFE), Université Paul-Valéry - Montpellier 3 (UPVM)-École Pratique des Hautes Études (EPHE), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro Montpellier, Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Université de Montpellier (UM), Computations were performed using HPC/AI resources from GENCI-IDRIS (Grant AD011012019R1)., ANR-10-LABX-0020,NUMEV,Digital and Hardware Solutions and Modeling for the Environement and Life Sciences(2010), and ANR-16-IDEX-0006,MUSE,MUSE(2016)
Subjects: [SDV.EE]Life Sciences [q-bio]/Ecology, environment
Abstract: The movement of animals is a central component of their behavioural strategies. Statistical tools for movement data analysis, however, have long been limited, and in particular, unable to account for past movement information except in a very simplified way. In this work, we propose MoveFormer, a new step-based model of movement capable of learning directly from full animal trajectories. While inspired by the classical step-selection framework and previous work on the quantification of uncertainty in movement predictions, MoveFormer also builds upon recent developments in deep learning, such as the Transformer architecture, allowing it to incorporate long temporal contexts. The model predicts an animal’s next movement step given its past movement history, including not only purely positional and temporal information, but also any available environmental covariates such as land cover or temperature. We apply our model to a diverse dataset made up of over 1550 trajectories from over 100 studies, and show how it can be used to gain insights about the importance of the provided context features, including the extent of past movement history. Our software, along with the trained model weights, is released as open source.
Published: 2023

11. Interrogation de bases de données biologiques publiques par reformulation de requêtes et classement des résultats avec ConQuR-Bio

Author: Brancotte, Bryan, Rance, Bastien, Denise, Alain, Cohen-Boulakia, Sarah, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Centre de Recherche des Cordeliers (CRC), Université Paris Diderot - Paris 7 (UPD7)-École pratique des hautes études (EPHE), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Université Pierre et Marie Curie - Paris 6 (UPMC)-Université Paris Descartes - Paris 5 (UPD5)-Institut National de la Santé et de la Recherche Médicale (INSERM), Scientific Data Management (ZENITH), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Modeling plant morphogenesis at different scales, from genes to phenotype (VIRTUAL PLANTS), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de la Recherche Agronomique (INRA)-Amélioration génétique et adaptation des plantes méditerranéennes et tropicales (UMR AGAP), Institut National de la Recherche Agronomique (INRA)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro), Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de la Recherche Agronomique (INRA)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro), Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Institut de Biologie Computationnelle (IBC), Université de Montpellier (UM)-Institut National de la Recherche Agronomique (INRA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS), Université Paris Diderot - Paris 7 (UPD7)-École pratique des hautes études (EPHE)-Université Pierre et Marie Curie - Paris 6 (UPMC)-Université Paris Descartes - Paris 5 (UPD5)-Institut National de la Santé et de la Recherche Médicale (INSERM), Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Inria Sophia Antipolis - Méditerranée (CRISAM), Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro)-Institut National de la Recherche Agronomique (INRA)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro), Amélioration génétique et adaptation des plantes méditerranéennes et tropicales (UMR AGAP), Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro)-Institut National de la Recherche Agronomique (INRA)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro)-Institut National de la Recherche Agronomique (INRA)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de la Recherche Agronomique (INRA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS), Laboratoire de Recherche en Informatique ( LRI ), Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ), Centre de Recherche des Cordeliers ( CRC ), Université Paris Diderot - Paris 7 ( UPD7 ) -École pratique des hautes études ( EPHE ) -Université Pierre et Marie Curie - Paris 6 ( UPMC ) -Université Paris Descartes - Paris 5 ( UPD5 ) -Institut National de la Santé et de la Recherche Médicale ( INSERM ), Scientific Data Management ( ZENITH ), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier ( LIRMM ), Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ) -Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ) -Inria Sophia Antipolis - Méditerranée ( CRISAM ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Institut National de Recherche en Informatique et en Automatique ( Inria ), Modeling plant morphogenesis at different scales, from genes to phenotype ( VIRTUAL PLANTS ), Inria Sophia Antipolis - Méditerranée ( CRISAM ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -Institut National de la Recherche Agronomique ( INRA ) -Centre de coopération internationale en recherche agronomique pour le développement [CIRAD] : UMR51, Institut de Biologie Computationnelle ( IBC ), Centre de Coopération Internationale en Recherche Agronomique pour le Développement ( CIRAD ) -Institut National de la Recherche Agronomique ( INRA ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -Université de Montpellier ( UM ) -Centre National de la Recherche Scientifique ( CNRS ), Université Pierre et Marie Curie - Paris 6 (UPMC)-École Pratique des Hautes Études (EPHE), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Université Paris Diderot - Paris 7 (UPD7)-Université Paris Descartes - Paris 5 (UPD5)-Institut National de la Santé et de la Recherche Médicale (INSERM), and Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de la Recherche Agronomique (INRA)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro)-Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de la Recherche Agronomique (INRA)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro)
Subjects: [ INFO.INFO-IR ] Computer Science [cs]/Information Retrieval [cs.IR], [ INFO.INFO-DB ] Computer Science [cs]/Databases [cs.DB], [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], [ INFO.INFO-BI ] Computer Science [cs]/Bioinformatics [q-bio.QM], [INFO.INFO-IR]Computer Science [cs]/Information Retrieval [cs.IR], [INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS], [INFO.INFO-BI]Computer Science [cs]/Bioinformatics [q-bio.QM], [ INFO.INFO-DS ] Computer Science [cs]/Data Structures and Algorithms [cs.DS]
Abstract: National audience; L’analyse d'expériences bioinformatiques comprend la comparaison des nouveaux résultats obtenus auxdonnées existantes. Durant ces trente dernières années, les scientifiques ont du faire face à une avalanche dedonnées, de différents types, et présentes dans une multitude de bases de données publiques. L’accès auxdonnées publiques se fait par l’interrogation de portails (tels que le portail Entrez du NCBI) au moyen demots clés. Cependant, deux requêtes très similaires peuvent fournir des ensembles de réponses différentsconduisant l’utilisateur à devoir tester différentes reformulations de ses requêtes (termes synonymes,variantes orthographiques, abréviations).... Les résultats obtenus doivent ensuite être filtrés, comparés... Enoutre, chaque ensemble de résultats est classé par le portail (en utilisant le nombre d'occurrences du mot-clédans chaque résultat). Cependant, lorsque plusieurs reformulations sont considérées, il n'est pas simple deproduire un classement triant par ordre de pertinence l'ensemble des résultats recueillis séparément, d'autantque ces résultats peuvent être fournis par centaines.Dans cette démonstration, nous présentons ConQuR-Bio (http://conqur-bio.lri.fr) qui permet auxutilisateurs d'interroger les bases de données publiques du NCBI tout en générant automatiquement toutes lesreformulations possibles et fournit des réponses triées en utilisant des techniques de consensus de classement(ou agrégation de classements). Notre démonstration montrera l’intérêt de notre approche pour des requêtesbiomédicales, lors de la recherche de gènes issus d’EntrezGene et impliqués dans des maladies.
Published: 2015

12. Two-Phase Scheduling for Efficient Vehicle Sharing

Author: Patrick Valduriez, Lei Mo, Ji Liu, Carlyna Bondiombouy, Baidu Research, African Institute for Mathematical Sciences (AIMS), Southeast University [Jiangsu], Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Inria Sophia Antipolis - Méditerranée (CRISAM), and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: Optimization, Mathematical optimization, Schedule, [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], Job shop scheduling, Scheduling, Total cost, Computer science, Mechanical Engineering, Scheduling (production processes), Context (language use), 7. Clean energy, Computer Science Applications, Path planning problem, SAFER, 11. Sustainability, Automotive Engineering, Path (graph theory), [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], Intelligent transportation system, Vehicle sharing
Abstract: International audience; Cooperative Intelligent Transport Systems (C-ITS) is a promising technology to make transportation safer and more efficient. Ridesharing for long-distance is becoming a key means of transportation in C-ITS. In this paper, we focus on private long-distance ridesharing, which reduces the total cost of vehicle utilization for long-distance journeys. In this context, we investigate journey scheduling problem with shared vehicles to reduce the total cost of vehicle utilization. Most of the existing works directly schedule journeys to vehicles with long scheduling time and only consider the cost of driving travellers instead of the total cost. In contrast, to reduce the total cost and scheduling time, we propose a comprehensive cost model and a two-phase journey scheduling approach, which includes path generation and path scheduling. On this basis, we propose two path generation methods: a simple near optimal method and a reset near optimal method as well as a greedy based path scheduling method. Finally, we present an experimental evaluation with different path generation and path scheduling methods with synthetic data generated based on real-world data. The results reveal that the proposed scheduling approach significantly outperforms baseline methods in terms of total cost (up to 69.8%) and scheduling time (up to 84.0%) and the scheduling time is reasonable (up to 0.16s). The results also show that our approach has higher efficiency (up to 141.7%) than baseline methods.
Published: 2022

13. Variable-Size Segmentation for Time Series Representation

Author: Djebour, Lamia, Akbarinia, Reza, Masseglia, Florent, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)
Subjects: Segmentation, Similarity Search, SAX, [INFO.INFO-IR]Computer Science [cs]/Information Retrieval [cs.IR], Information Retrieval, Time Series, Representation
Abstract: International audience; Given the high data volumes in time series applications, or simply the need for fast response times, it is usually necessary to rely on alternative, shorter representations of time series, usually with information loss. This incurs approximate comparisons of time series where precision is a major issue. We propose a new representation approach called ASAX, coming with two techniques ASAX EN and ASAX SAE, for segmenting time series before their transformation into symbolic representations. Our solution can reduce significantly the error incurred by possible splittings at different steps of the representation calculation, by taking into account the entropy of the representations (ASAX EN) or the sum of absolute errors (ASAX SAE), particularly for datasets with unbalanced (non-uniform) distributions. This is particularly useful for time series similarity search, which is the core of many data analytics tasks. We provide theoretical guarantees on the lower bound of similarity measures, and our experiments illustrate that our approach can improve significantly the time series representation quality.
Published: 2023

14. ProvLight: Efficient Workflow Provenance Capture on the Edge-to-Cloud Continuum

Author: Rosendo, Daniel, Mattoso, Marta, Costan, Alexandru, Souza, Renan, Pina, Débora, Valduriez, Patrick, Antoniu, Gabriel, Scalable Storage for Clouds and Beyond (KerData), Inria Rennes – Bretagne Atlantique, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-SYSTÈMES LARGE ÉCHELLE (IRISA-D1), Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-IMT Atlantique (IMT Atlantique), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT)-Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-IMT Atlantique (IMT Atlantique), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de Engenharia (COPPE-UFRJ), Universidade Federal do Rio de Janeiro (UFRJ), Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA), Oak Ridge National Laboratory [Oak Ridge] (ORNL), UT-Battelle, LLC, HPC-BigData Inria Challenge (IPL), and ANR-15-CE25-0003,OverFlow,Workflow Data Management as a Service pour des Applications Multi-Site(2015)
Subjects: FOS: Computer and information sciences, IoT, [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], Computer Science - Performance, Databases (cs.DB), Computing Continuum, Workflows, Performance (cs.PF), [INFO.INFO-PF]Computer Science [cs]/Performance [cs.PF], Computer Science - Databases, Computer Science - Distributed, Parallel, and Cluster Computing, Lineage, Edge, Provenance, Distributed, Parallel, and Cluster Computing (cs.DC), [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC]
Abstract: Modern scientific workflows require hybrid infrastructures combining numerous decentralized resources on the IoT/Edge interconnected to Cloud/HPC systems (aka the Computing Continuum) to enable their optimized execution. Understanding and optimizing the performance of such complex Edge-to-Cloud workflows is challenging. Capturing the provenance of key performance indicators, with their related data and processes, may assist in understanding and optimizing workflow executions. However, the capture overhead can be prohibitive, particularly in resource-constrained devices, such as the ones on the IoT/Edge.To address this challenge, based on a performance analysis of existing systems, we propose ProvLight, a tool to enable efficient provenance capture on the IoT/Edge. We leverage simplified data models, data compression and grouping, and lightweight transmission protocols to reduce overheads. We further integrate ProvLight into the E2Clab framework to enable workflow provenance capture across the Edge-to-Cloud Continuum. This integration makes E2Clab a promising platform for the performance optimization of applications through reproducible experiments.We validate ProvLight at a large scale with synthetic workloads on 64 real-life IoT/Edge devices in the FIT IoT LAB testbed. Evaluations show that ProvLight outperforms state-of-the-art systems like ProvLake and DfAnalyzer in resource-constrained devices. ProvLight is 26 -- 37x faster to capture and transmit provenance data; uses 5 -- 7x less CPU; 2x less memory; transmits 2x less data; and consumes 2 -- 2.5x less energy. ProvLight and E2Clab are available as open-source tools.
Published: 2023
Full Text: View/download PDF

15. Black-box language model explanation by context length probing

Author: Cífka, Ondřej, Liutkus, Antoine, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), ANR-10-LABX-0020,NUMEV,Digital and Hardware Solutions and Modeling for the Environement and Life Sciences(2010), ANR-16-IDEX-0006,MUSE,MUSE(2016), Cífka, Ondřej, Laboratoires d'excellence - Digital and Hardware Solutions and Modeling for the Environement and Life Sciences - - NUMEV2010 - ANR-10-LABX-0020 - LABX - VALID, and MUSE - - MUSE2016 - ANR-16-IDEX-0006 - IDEX - VALID
Subjects: FOS: Computer and information sciences, [INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI], Computer Science - Machine Learning, Computer Science - Computation and Language, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], [INFO.INFO-CL] Computer Science [cs]/Computation and Language [cs.CL], [INFO.INFO-LG] Computer Science [cs]/Machine Learning [cs.LG], Computation and Language (cs.CL), [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], Machine Learning (cs.LG), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract: The increasingly widespread adoption of large language models has highlighted the need for improving their explainability. We present context length probing, a novel explanation technique for causal language models, based on tracking the predictions of a model as a function of the length of available context, and allowing to assign differential importance scores to different contexts. The technique is model-agnostic and does not rely on access to model internals beyond computing token-level probabilities. We apply context length probing to large pre-trained language models and offer some initial analyses and insights, including the potential for studying long-range dependencies. The source code and an interactive demo of the method are available., 11 pages, 9 figures. ACL 2023 short paper camera-ready. Demos at https://cifkao.github.io/context-probing/ and https://huggingface.co/spaces/cifkao/context-probing ; code at https://github.com/cifkao/context-probing/
Published: 2022

16. Interprétabilité des modèles de distribution d’espèces basés sur des réseaux de neurones convolutifs

Author: Deneu, Benjamin, Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Univeristé de Montpellier, Alexis Joly, and François Munoz
Subjects: Réseaux de neurones convolutif, [SDE]Environmental Sciences, Réseaux de neurones, Convolutional neural network, Modelisation, Deep neural network, Distribution des espèces, Species distribution, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract: Species distribution models link the geographic distribution of a species to its environment. The objectives of using these models are multiple. They can be used to extract knowledge on species and their environmental preferences, to help with conservation plans and policies, to monitor and anticipate the spread of invasive species, or to simulate environmental changes and their impacts on species. To best meet these objectives, it is necessary to design efficient, accurate and interpretable models. Most of the models used today are relatively simple models. These models have the advantage of being easy to interpret by producing simple relationships between a species and its environment. However, they often share some shortcomings such as sensitivity to overfitting, which requires a careful choice of descriptive data of the environment to avoid interpretation errors. Models based on machine learning approaches have shown performances that are often as good or even better, with a stronger robustness against overfitting. However, these methods are more often criticized for their lack of interpretability. This is the case with convolutional neural networks whose first experiments have shown promising results for their use in species distribution modeling. Convolutional neural networks are known for their particularly high performance in all image processing tasks (classification, object detection, counting, etc.). They have the particularity of being able to use very large data with little risk of overfitting. Even more than other machine learning approaches, these models are often described as black boxes that are difficult to interpret. We propose to study the use of these models, called Deep-SDMs, in the context of species distribution prediction with a particular attention to interpretation in order to highlight the potential interests of this new approach while trying to clarify the mechanisms involved.We present the use and analysis of Deep-SDMs with several interpretability experiments in different contexts. We conduct comparisons on some aspects with more state-of-the-art models. We propose qualitative and quantitative analyses on the interpretation of Deep-SDMs learning. In particular, we propose to study what the model captures, either by analyzing the differences in performance according to the data used and the information they contain, or directly by studying the learned representation space of the model (the last layer of the model).Overall we show that it is possible to analyze and interpret model learning in several ways, leading to interesting ecological conclusions. We show an interesting potential of Deep-SDMs that allow: (1) to learn a single model for many species simultaneously and using observation data without absence data, (2) to use more complex and richer representations of the environment thanks to their ability to use very high dimensional data, (3) often better performances than other models, especially on rare species, (4) learning on a very large scale (thousands of species and regions the size of countries) and at a very fine resolution (around ten meters) thanks to remote sensing data, and (5) possible reuse of models in similar contexts, taking advantage of the learning already done.; Les modèles de distributions d'espèces font le lien entre la distribution géographique d’une espèce et son environnement. Les objectifs de l’utilisation de ces modèles sont multiples. On peut citer entre autres l’extraction de connaissance sur les espèces et leur préférences environnementales, l’aide aux plans de conservations et politiques de protections des espèces, la surveillance et l’anticipation de la propagation d’espèces envahissantes ou encore les simulations d’évolution de l’environnement et leurs impacts sur les espèces. Pour répondre au mieux à ces objectifs il est nécessaire de concevoir des modèles performants, précis et interprétables. La plupart des modèles utilisés aujourd’hui sont des modèles relativement peu complexes. Ces modèles ont l’avantage d’être faciles à interpréter en produisant des relations simples entre une espèce et son environnement. Cependant, ils partagent souvent certains défauts comme la sensibilité au sur-apprentissage nécessitant ainsi de bien choisir les données descriptives de l’environnement pour éviter les erreurs d’interprétation. Des modèles basés sur des approches d’apprentissage artificiel ont montré des performances souvent aussi bonnes voire meilleures avec une plus forte robustesse contre le sur-apprentissage. Ces méthodes sont en revanche plus souvent critiquées pour leur manque d’interprétabilité. C'est le cas avec les réseaux de neurones convolutifs dont les premières expériences ont montré des résultats prometteurs pour leur utilisation en modélisation de la distribution d’espèces. Les réseaux de neurones convolutifs sont connus pour leurs performances particulièrement élevées dans toutes les tâches de traitement d’image (classification, détection d’objets, comptage, etc.). Ils ont la particularité de pouvoir utiliser des données de très grandes dimensions avec peu de risques de sur-apprentissage. Encore plus que les autres modèles d’apprentissage artificiel, ces modèles sont très souvent décrits comme des boîtes noires difficiles à interpréter. Nous proposons d’étudier l’utilisation de ces modèles, appelés Deep-SDMs, dans le contexte de la prédiction de distribution d’espèces en portant une attention particulière à l'interprétation afin de mettre en avant les intérêts potentiels de cette nouvelle approche tout en essayant d’éclaircir au mieux les mécanismes en jeu.Nous présentons l’utilisation et l’analyse des modèles Deep-SDMs avec plusieurs expériences d’interprétabilité dans différents contextes. Nous menons des comparaisons sur certains aspects avec des modèles plus état-de-l’art. Nous proposons des analyses qualitatives et quantitatives sur l’interprétation des apprentissages de Deep-SDMs. Nous proposons notamment d’étudier ce que le modèle capture, soit en analysant les différences de performances selon les données utilisées et les informations qu’elles contiennent, soit directement en étudiant l’espace de représentation appris du modèle (la dernière couche du modèle).Dans l’ensemble nous montrons qu’il est possible d'analyser et d'interpréter l’apprentissage des modèles de plusieurs manières, permettant d’aboutir à des conclusions écologiques intéressantes. Nous montrons un potentiel intéressant des Deep-SDMs qui permettent: (1) d’apprendre un unique modèles pour de nombreuses espèces simultanément et en utilisant des données d’observations sans données d’absences, (2) d’utiliser des représentations plus complexes et plus riches de l’environnement grâce à leur capacité à utiliser des données de très grande dimension, (3) des performances souvent meilleures que les autres modèles, en particulier sur les espèces rares, (4) un apprentissage à la fois à très grande échelle (sur des milliers d’espèces et des régions de la taille de pays) et à très fine résolution (de l’ordre de la dizaine de mètres) grâce aux données de remote sensing, et (5) une réutilisation possible des modèles dans des contextes proches profitant en partie des apprentissage déjà effectués.
Published: 2022

17. StreamCloud: An Elastic Parallel-Distributed Stream Processing Engine

Author: Gulisano, Vincenzo, Jiménez Peris, Ricardo, Valduriez, Patrick, Distributed Systems Laboratory (DSL), Universidad Politécnica de Madrid (UPM), Scientific Data Management (ZENITH), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Universidad Politécnica de Madrid, Ricardo Jiménez Peris, Patrick Valduriez(Patrick.Valduriez@inria.fr), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Inria Sophia Antipolis - Méditerranée (CRISAM)
Subjects: Informática, Telecomunicaciones, Elasticidad, Scalability, Fault Tolerance, Stream Processing Engine, Data Streaming, Elasticity, Load Balancing, Sistemas de procesamiento de flujos de datos, Equilibrado de Carga, Tolerancia a fallos, [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], Escalabilidad, Elasticité
Abstract: In recent years, applications in domains such as telecommunications, network security or large scale sensor networks showed the limits of the traditional store-then-process paradigm. In this context, Stream Processing Engines emerged as a candidate solution for all these applications demanding for high processing capacity with low processing latency guarantees. With Stream Processing Engines, data streams are not persisted but rather processed on the fly, producing results continuously. Current Stream Processing Engines, either centralized or distributed, do not scale with the input load due to single-node bottlenecks. Moreover, they are based on static configurations that lead to either under or over-provisioning. This Ph.D. thesis discusses StreamCloud, an elastic paralleldistributed stream processing engine that enables for processing of large data stream volumes. Stream- Cloud minimizes the distribution and parallelization overhead introducing novel techniques that split queries into parallel subqueries and allocate them to independent sets of nodes. Moreover, Stream- Cloud elastic and dynamic load balancing protocols enable for effective adjustment of resources depending on the incoming load. Together with the parallelization and elasticity techniques, Stream- Cloud defines a novel fault tolerance protocol that introduces minimal overhead while providing fast recovery. StreamCloud has been fully implemented and evaluated using several real word applications such as fraud detection applications or network analysis applications. The evaluation, conducted using a cluster with more than 300 cores, demonstrates the large scalability, the elasticity and fault tolerance effectiveness of StreamCloud.; En los útimos años, aplicaciones en dominios tales como telecomunicaciones, seguridad de redes y redes de sensores de gran escala se han encontrado con múltiples limitaciones en el paradigma tradicional de bases de datos. En este contexto, los sistemas de procesamiento de flujos de datos han emergido como solución a estas aplicaciones que demandan una alta capacidad de procesamiento con una baja latencia. En los sistemas de procesamiento de flujos de datos, los datos no se persisten y luego se procesan, en su lugar los datos son procesados al vuelo en memoria produciendo resultados de forma continua. Los actuales sistemas de procesamiento de flujos de datos, tanto los centralizados, como los distribuidos, no escalan respecto a la carga de entrada del sistema debido a un cuello de botella producido por la concentración de flujos de datos completos en nodos individuales. Por otra parte, éstos están basados en configuraciones estáticas lo que conducen a un sobre o bajo aprovisionamiento. Esta tesis doctoral presenta StreamCloud, un sistema elástico paralelo-distribuido para el procesamiento de flujos de datos que es capaz de procesar grandes volúmenes de datos. StreamCloud minimiza el coste de distribución y paralelización por medio de una técnica novedosa la cual particiona las queries en subqueries paralelas repartiéndolas en subconjuntos de nodos independientes. Ademas, Stream- Cloud posee protocolos de elasticidad y equilibrado de carga que permiten una optimización de los recursos dependiendo de la carga del sistema. Unidos a los protocolos de paralelización y elasticidad, StreamCloud define un protocolo de tolerancia a fallos que introduce un coste mínimo mientras que proporciona una rápida recuperación. StreamCloud ha sido implementado y evaluado mediante varias aplicaciones del mundo real tales como aplicaciones de detección de fraude o aplicaciones de análisis del tráfico de red. La evaluación ha sido realizada en un cluster con más de 300 núcleos, demostrando la alta escalabilidad y la efectividad tanto de la elasticidad, como de la tolerancia a fallos de StreamCloud.; In recent years...
Published: 2022

18. Identify ambiguous tasks combining crowdsourced labels by weighting Areas Under the Margin

Author: Lefort, Tanguy, Charlier, Benjamin, Joly, Alexis, Salmon, Joseph, Institut Montpelliérain Alexander Grothendieck (IMAG), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Institut Universitaire de France (IUF), Ministère de l'Education nationale, de l’Enseignement supérieur et de la Recherche (M.E.N.E.S.R.), and ANR-20-CHIA-0001,CAMELOT,Apprentissage automatique et optimisation coopératifs.(2020)
Subjects: FOS: Computer and information sciences, Computer Science - Machine Learning, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Statistics - Machine Learning, I.2.6, Computer Science - Human-Computer Interaction, Machine Learning (stat.ML), [STAT.OT]Statistics [stat]/Other Statistics [stat.ML], Machine Learning (cs.LG), Human-Computer Interaction (cs.HC), 68T05
Abstract: In supervised learning - for instance in image classification - modern massive datasets are commonly labeled by a crowd of workers. The obtained labels in this crowdsourcing setting are then aggregated for training. The aggregation step generally leverages a per-worker trust score. Yet, such worker-centric approaches discard each task's ambiguity. Some intrinsically ambiguous tasks might even fool expert workers, which could eventually be harmful to the learning step. In a standard supervised learning setting - with one label per task - the Area Under the Margin (AUM) is tailored to identify mislabeled data. We adapt the AUM to identify ambiguous tasks in crowdsourced learning scenarios, introducing the Weighted AUM (WAUM). The WAUM is an average of AUMs weighted by task-dependent scores. We show that the WAUM can help discard ambiguous tasks from the training set, leading to better generalization or calibration performance. We report improvements over existing strategies for learning a crowd, both for simulated settings and for the CIFAR-10H, LabelMe and Music crowdsourced datasets.
Published: 2022

19. A Data-Driven Model Selection Approach to Spatio-Temporal Prediction

Author: Rocío Zorrilla, Eduardo Ogasawara, Patrick Valduriez, Fábio Porto, Laboratorio Nacional de Computação Cientifica [Rio de Janeiro] (LNCC / MCT), Centro Federal de Educação Tecnológica Celso Suckow da Fonseca (Rio de Janeiro) ( CEFET/RJ), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), SBBD, Sociedade Brazileira de Computação (SBC), and équipe associée Inria hpdasc
Subjects: [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], Spatio-temporal Predictive Queries
Abstract: XXXVII WORKSHOP BRASILEIRO DE INTEGRAÇÃO DE BANCOS DE DADOS E INTELIGÊNCIA ARTIFICIAL; International audience; Spatio-temporal Predictive Queries encompass a spatio-temporal constraint, defining a region, a target variable, and an evaluation metric. The output of such queries presents the future values for the target variable computed by predictive models at each point of the spatio-temporal region. Unfortunately, especially for large spatio-temporal domains with millions of points, training temporal models at each spatial domain point is prohibitive. In this work, we propose a data-driven approach for selecting pre-trained temporal models to be applied at each query point. The chosen approach applies a model to a point according to the training and input time series similarity. The approach avoids training a different model for each domain point, saving model training time. Moreover, it provides a technique to decide on the best-trained model to be applied to a point for prediction. In order to assess the applicability of the proposed strategy, we evaluate a case study for temperature forecasting using historical data and auto-regressive models. Computational experiments show that the proposed approach, compared to the baseline, achieves equivalent predictive performance using a composition of pre-trained models at a fraction of the total computational cost.
Published: 2022

20. Integrating Machine Learning Model Ensembles to the SAVIME Database System

Author: Silva, Anderson, Valduriez, Patrick, Machado Porto, Fábio André, Laboratorio Nacional de Computação Cientifica [Rio de Janeiro] (LNCC / MCT), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Sociedade Brazileira de Computação (SBC), équipe associée Inria hpdasc, and European Project: 101016478,RISC2
Subjects: Machine Learning, Data Management Systems, [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], Database systems, Machine learning models, Ensemble Classifier, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract: XXXVII WORKSHOP BRASILEIRO DE INTEGRAÇÃO DE BANCOS DE DADOS E INTELIGÊNCIA ARTIFICIAL; International audience; The integration of machine learning algorithms into database systems has brought new opportunities in different areas from indexing to query optimization. In this paper, we describe the integration of an approach for the automatic computation of model ensembles to answer a predictive query. We have extended the SAVIME multi-dimensional array DBMS by adding a new function to its query language and implementing the selection and allocation ensemble model dataflow into the query processing component of SAVIME. We show some initial experimental results depicting its performance against a pure Python implementation of the ensemble approach. Interestingly enough the C++ implementation within SAVIME is up to 4 times faster than its competitor.
Published: 2022

21. Techniques de segmentation adaptative pour une représentation efficace des séries temporelles

Author: Djebour, Lamia, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Université de Montpellier, Florent Masseglia, and Reza Akbarinia
Subjects: Representations, Time series, Segmentation, Similarity Search, Erreur d’approximation, Approximation Error, Information Retrieval, Séries temporelles, [INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS], Représentations, Recherche de similarité, Extraction d'informations, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL]
Abstract: Many applications in different domains generate time series data at an increasing rate. The continuous flow of emitted data may concern personal activities (e.g., through smart-meters or smart-plugs for electricity or water consumption) or professional activities (e.g., for monitoring heart activity or through the sensors installed on plants by farmers). This results in the production of large and complex data, usually in the form of time series.In recent years, there has been an explosion of interest in time series data mining. As a general rule, large time series come along with super-high dimensionality. As a consequence, it is difficult and inefficient to directly mine the raw time series without relying on dimensionality reduction. Therefore, the representation of the data is the key to efficient and effective solutions. Given this high data volumes in time series applications, or simply the need for fast response times, it is usually necessary to rely on alternative, shorter representations of these series, usually with loss. This incurs approximate comparisons of time series where precision is a major issue.In this thesis, we focus on the problem of segmenting time series before their transformation into symbolic representations. For this, we propose solutions to adaptively segment time series databases by adopting a variable segment size that depends on the time series distribution. These methods reduce significantly the information loss incurred by possible splittings at different steps of the representation calculation, particularly for datasets with unbalanced (non-uniform) distributions.We provide theoretical guarantees on the lower bound of similarity measures, and our results illustrate that our techniques can significantly improve the time series representation quality.; De nombreuses applications dans différents domaines génèrent des données de séries temporelles à un rythme croissant. Le flux continu de données émises peut concerner des activités personnelles (par exemple, au moyen de compteurs intelligents ou de prises connectées pour la consommation d’électricité ou d’eau) ou professionnelles (par exemple, pour la surveillance de l’activité cardiaque ou à travers les capteurs installés sur les plantes par les agriculteurs). Il en résulte une production de données volumineuses et complexes, généralement sous la forme de séries temporelles.Généralement, les bases de données de séries temporelles sont caractérisées par leur très grand volume. Par conséquent, il est difficile et inefficace d’exploiter directement les données de séries temporelles brutes sans avoir recours à la réduction de la dimensionnalité.Ce verrou motive l’étude de représentations représentations alternatives, plus courtes, qui résument les séries d’origine avec une perte d’information acceptable. Les comparaisons de séries temporelles qui se basent sur ces représentations sont alors approximatives, ce qui fait de la précision un enjeu majeur.Dans cette thèse, nous étudions le problème de la segmentation des séries temporelles avant qu’elles soient transformées en représentations symboliques. Pour cela, nous proposons des solutions de segmentation adaptative des séries temporelles en adoptant une taille de segment variable qui dépend de la distribution de ces séries. Ces méthodes réduisent de manière significative la perte d’information due aux découpages possibles dans les différentes étapes du calcul de la représentation, en particulier pour les ensembles de données dont les distributions sont non uniformes.Nous fournissons des garanties théoriques sur la borne inférieure des mesures de similitude entre séries temporelles, et nos résultats montrent que nos techniques peuvent améliorer considérablement la qualité de la représentation des séries temporelles.
Published: 2022

22. Overview of LifeCLEF 2022: An Evaluation of Machine-Learning Based Species Identification and Species Distribution Prediction

Author: Alexis Joly, Hervé Goëau, Stefan Kahl, Lukáš Picek, Titouan Lorieul, Elijah Cole, Benjamin Deneu, Maximilien Servajean, Andrew Durso, Hervé Glotin, Robert Planqué, Willem-Pier Vellinga, Amanda Navine, Holger Klinck, Tom Denton, Ivan Eggel, Pierre Bonnet, Milan Šulc, Marek Hrúz, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), California Institute of Technology (CALTECH), ADVanced Analytics for data SciencE (ADVANSE), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Université Paul-Valéry - Montpellier 3 (UPVM), Laboratoire d'Informatique et Systèmes (LIS), Aix Marseille Université (AMU)-Université de Toulon (UTLN)-Centre National de la Recherche Scientifique (CNRS), Xeno-canto foundation, University of Hawai'i [Hilo], Google LLC, University of West Bohemia [Plzeň ], Alberto Barrón-Cedeño, Giovanni Da San Martino, Mirko Degli Esposti, Fabrizio Sebastiani, Craig Macdonald, Gabriella Pasi, Allan Hanbury, Martin Potthast, Guglielmo Faggioli, Nicola Ferro, and European Project: 863463,Cos4Cloud project
Subjects: Large-scales, Living organisms, Biodiversity monitoring, [SDV.BV.BOT]Life Sciences [q-bio]/Vegetal Biology/Botanics, Biodiversity conservation, Species Distribution Prediction, [SDV.BID.SPT]Life Sciences [q-bio]/Biodiversity/Systematics, Phylogenetics and taxonomy, Species distributions, [SDV.EE.ECO]Life Sciences [q-bio]/Ecology, environment/Ecosystems, Machine learning, LifeCLEF 2022, Monitoring system, Geographics, [SDE.BE]Environmental Sciences/Biodiversity and Ecology, Machine-learning, Species identification, General publics
Abstract: International audience; Building accurate knowledge of the identity, the geographic distribution and the evolution of species is essential for the sustainable development of humanity, as well as for biodiversity conservation. However, the difficulty of identifying plants, animals and fungi is hindering the aggregation of new data and knowledge. Identifying and naming living organisms is almost impossible for the general public and is often difficult even for professionals and naturalists. Bridging this gap is a key step towards enabling effective biodiversity monitoring systems. The LifeCLEF campaign, presented in this paper, has been promoting and evaluating advances in this domain since 2011. The 2022 edition proposes five data-oriented challenges related to the identification and prediction of biodiversity: (i) PlantCLEF: very large-scale plant identification, (ii) BirdCLEF: bird species recognition in audio soundscapes, (iii) GeoLifeCLEF: remote sensing based prediction of species, (iv) SnakeCLEF: snake species identification on a global scale, and (v) FungiCLEF: fungi recognition as an open set classification problem. This paper overviews the motivation, methodology and main outcomes of that five challenges.
Published: 2022

23. Species Distribution Modeling based on aerial images and environmental features with Convolutional Neural Networks

Author: Leblanc, César, Joly, Alexis, Lorieul, Titouan, Servajean, Maximilien, Bonnet, Pierre, Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), ADVanced Analytics for data SciencE (ADVANSE), Université Paul-Valéry - Montpellier 3 (UPVM), Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Guglielmo Faggioli, Nicola Ferro, Allan Hanbury, and Martin Potthast
Subjects: Presenceonly data, [SDV.EE.ECO]Life Sciences [q-bio]/Ecology, environment/Ecosystems, Aerial images, Species distribution modelling, Biodiversity, LifeCLEF, [SDE.BE]Environmental Sciences/Biodiversity and Ecology, [SDV.BV.BOT]Life Sciences [q-bio]/Vegetal Biology/Botanics, [SDV.BID.SPT]Life Sciences [q-bio]/Biodiversity/Systematics, Phylogenetics and taxonomy, [INFO.INFO-SI]Computer Science [cs]/Social and Information Networks [cs.SI], Environmental features
Abstract: International audience; Predicting which species are likely to be observed at a given location is an important issue both from a scientific point of view and for citizens interested in biodiversity. The aim of the GeoLifeCLEF challenge is to predict the presence of around 17K plant and animal species using 1.6M geo-localized observations from France and the US. Beyond GPS coordinates, additional covariates are provided for each observation: remote sensing imagery, land cover data, altitude data, bioclimatic data and pedologic data. We tested two simple approaches making use of every covariate available: (i) training separate models from them, testing them separately, and, (ii) combining those models together in the simplest manner, i.e., averaging their predicted scores. These simple methods allowed us to reach the fourth position on the private leaderboard of the competition. We also managed to improve our score after the end of the challenge by implementing a multi-modal convolutional neural network with separated features extractors.
Published: 2022

24. Overview of BirdCLEF 2022: Endangered bird species recognition in soundscape recordings

Author: Kahl, Stefan, Navine, Amanda, Denton, Tom, Klinck, Holger, Hart, Patrick, Glotin, Hervé, Goëau, Hervé, Vellinga, Willem-Pier, Planqué, Robert, Joly, Alexis, Cornell University [New York], University of Hawai'i [Hilo], Google LLC, Cornell Laboratory of Ornithology, Université de Toulon (UTLN), Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Xeno-canto foundation, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)
Subjects: bird, evaluation, call, species, [SDV.BV.BOT]Life Sciences [q-bio]/Vegetal Biology/Botanics, [SDV.BID.SPT]Life Sciences [q-bio]/Biodiversity/Systematics, Phylogenetics and taxonomy, passive acoustic monitoring, bioacoustics, benchmark, [SDV.EE.ECO]Life Sciences [q-bio]/Ecology, environment/Ecosystems, audio, collection, identification, song, LifeCLEF, fine-grained classification, [SDE.BE]Environmental Sciences/Biodiversity and Ecology, retrieval
Abstract: International audience; As the "extinction capital of the world", Hawai'i has lost 68% of its native bird species, the consequences of which can harm entire ecosystems. With physical monitoring difficult, scientists have turned to sound recordings, as this approach could provide a passive, low labor, and cost-effective strategy for monitoring endangered bird populations. Current methods for processing large bioacoustic datasets involve manual review of each recording. This requires specialized training and prohibitively large amounts of time. Recent advances in machine learning have made it possible to automatically identify bird songs for common species with ample training data. However, it remains challenging to develop such tools for rare and endangered species. The main goal of the 2022 edition of BirdCLEF was to advance automated detection of rare and endangered bird species that lack large amounts of training data. The competition challenged participants to develop reliable analysis frameworks to detect and identify the vocalizations of rare bird species in continuous Hawaiian soundscapes utilizing limited training data.
Published: 2022

25. Overview of GeoLifeCLEF 2022: Predicting species presence from multi-modal remote sensing, bioclimatic and pedologic data

Author: Lorieul, Titouan, Cole, Elijah, Deneu, Benjamin, Servajean, Maximilien, Bonnet, Pierre, Joly, Alexis, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), California Institute of Technology (CALTECH), ADVanced Analytics for data SciencE (ADVANSE), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Université Paul-Valéry - Montpellier 3 (UPVM), Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Département Systèmes Biologiques (Cirad-BIOS), and Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)
Subjects: evaluation, multi-modal data, [SDV.BV.BOT]Life Sciences [q-bio]/Vegetal Biology/Botanics, [SDV.BID.SPT]Life Sciences [q-bio]/Biodiversity/Systematics, Phylogenetics and taxonomy, benchmark, [SDV.EE.ECO]Life Sciences [q-bio]/Ecology, environment/Ecosystems, remote sensing imagery, environmental data, species distribution, LifeCLEF, presence-only data, [SDE.BE]Environmental Sciences/Biodiversity and Ecology, species distribution models, biodiversity
Abstract: International audience; Understanding the geographic distribution of species is a key concern in conservation. By pairing species occurrences with environmental features, researchers can model the relationship between an environment and the species which may be found there. To advance research in this area, a large-scale machine learning competition called GeoLifeCLEF 2022 was organized. It relied on a dataset of 1.6 million observations from 17K species of animals and plants. These observations were paired with high-resolution remote sensing imagery, land cover data, and altitude, in addition to traditional lowresolution climate and soil variables. The main goal of the challenge was to better understand how to leverage remote sensing data to predict the presence of species at a given location. This paper presents an overview of the competition, synthesizes the approaches used by the participating groups, and analyzes the main results. In particular, we highlight the ability of remote sensing imagery and convolutional neural networks to improve predictive performance, complementary to traditional approaches.
Published: 2022

26. PeTriBERT : Augmenting BERT with tridimensional encoding for inverse protein folding and design

Author: Baldwin Dumortier, Antoine Liutkus, Clément Carré, Gabriel Krouk, Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Bionomeex [Montpellier], Institut des Sciences des Plantes de Montpellier (IPSIM), Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro Montpellier, and Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Université de Montpellier (UM)
Subjects: [SDV]Life Sciences [q-bio], [SDV.BV]Life Sciences [q-bio]/Vegetal Biology, transformers, AlphaFold, NLP, proteins, protein generation, inverse folding
Abstract: Protein is biology workhorse. Since the recent break-through of novel folding methods, the amount of available structural data is increasing, closing the gap between data-driven sequence-based and structure-based methods. In this work, we focus on the inverse folding problem that consists in predicting an amino-acid primary sequence from protein 3D structure. For this purpose, we introduce a simple Transformer model from Natural Language Processing augmented 3D-structural data. We call the resulting model PeTriBERT: Proteins embedded in tridimensional representation in a BERT model. We train this small 40-million parameters model on more than 350 000 proteins sequences retrieved from the newly available AlphaFoldDB database. Using PetriBert, we are able to in silico generate totally new proteins with a GFP-like structure. These 9 of 10 of these GFP structural homologues have no ressemblance when blasted on the whole entry proteome database. This shows that PetriBert indeed capture protein folding rules and become a valuable tool for de novo protein design.
Published: 2022

27. Data and Machine Learning Model Management with Gypscie

Author: Porto, F., Valduriez, P., Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Laboratorio Nacional de Computação Cientifica [Rio de Janeiro] (LNCC / MCT), SCALAC, Advanced Computing System for Latin America and the Caribbean (SCALAC), Inria associated team HPPDASC, and European Project: 101016478,RISC2
Subjects: MLSystems, Machine Learning, [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], Model Selection, Array DBMS, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract: International audience; As predictive analytics using ML models (or models for short) become preva- lent in different stages of scientific exploration, a new set of artifacts are pro- duced during the models’ life-cycle that need to be managed [2]. In addition to the models with their evolving versions, ML life-cycle artifacts include the collected training data and pre-processing workflows, data labels and selected features, model training, tuning and monitoring statistics and provenance in- formation. However, to realize the full potential of data science, these artifacts must be built and combined, which can be very complex as there can be many to select from. Furthermore, they should be shared and reused, in particular, in different execution environments such as HPC or Spark clusters. In order to support the complete ML life-cycle process and produced arti- facts, we have been developing the Gypscie framework, which offers collaborat- ing researchers a common software infrastructure to develop, share, improve and publish ML artifacts.
Published: 2022
Full Text: View/download PDF

28. Forward and Backward Inertial Anomaly Detector: A Novel Time Series Event Detection Method

Author: Janio Lima, Rebecca Salles, Fabio Porto, Rafaelli Coutinho, Pedro Alpis, Luciana Escobar, Esther Pacitti, Eduardo Ogasawara, Centro Federal de Educação Tecnológica Celso Suckow da Fonseca (Rio de Janeiro) ( CEFET/RJ), Laboratorio Nacional de Computação Cientifica [Rio de Janeiro] (LNCC / MCT), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), IEEE, and Equipe associée Inria HPDaSc
Subjects: [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], anomalies, inertia, time series, event detection
Abstract: International audience; Time series event detection is related to studying methods for detecting observations in a series with special meaning. These observations differ from the expected behavior of the data set. In data streaming scenarios, it is possible to observe an increase in the speed of data generation in time series. Therefore, adapting to time series changes becomes crucial. Thus, identifying events associated with these changes is essential for timely and correct decision-making. Although there are many methods to detect events, it is still possible to have difficulties detecting them correctly, particularly those associated with concept drift. In order to fill the gap in the literature, this work proposes a new method, named Forward and Backward Inertial Anomaly Detector (FBIAD), for detecting events in time series. It contributes by analyzing surrounding inertia around observations. FBIAD outperformed other methods both in accuracy and elapsed time.
Published: 2022

29. High-throughput and automatic structural and developmental root phenotyping on Arabidopsis seedlings

Author: Romain Fernandez, Amandine Crabos, Morgan Maillard, Philippe Nacry, Christophe Pradal, Amélioration génétique et adaptation des plantes méditerranéennes et tropicales (UMR AGAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro Montpellier, Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Université de Montpellier (UM), Institut des Sciences des Plantes de Montpellier (IPSIM), Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro Montpellier, Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)
Subjects: Root system architecture, Tracking, [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV], High-throughput phenotyping, Dynamic analysis, [SDV.BV]Life Sciences [q-bio]/Vegetal Biology, High-throughput phenotyping Computer vision Dynamic analysis Root system architecture Tracking, Computer vision, [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation
Abstract: BackgroundHigh-throughput phenotyping is crucial for the genetic and molecular understanding of adaptive root system development. In recent years, imaging automata have been developed to acquire the root system architecture of many genotypes grown in Petri dishes to explore the Genetic x Environment (GxE) interaction. There is now an increasing interest in understanding the dynamics of the adaptive responses, such as the organ apparition or the growth rate. However, due to the increasing complexity of root architectures in development, the accurate description of the topology, geometry, and dynamics of a growing root system remains a challenge.ResultsWe designed a high-throughput phenotyping method, combining an imaging device and an automatic analysis pipeline based on registration and topological tracking, capable of accurately describing the topology and geometry of observed root systems in 2D+t. The method was tested on a challenging Arabidopsis seedling dataset, including numerous root occlusions and crossovers. Static phenes are estimated with high accuracy (R2 = 0.996 and 0, 923 for primary and second-order roots length, respectively). These performances are similar to state-of-the-art results obtained on root systems of equal or lower complexity. In addition, our pipeline estimates dynamic phenes accurately between two successive observations (R2 = 0. 938 for lateral root growth).ConclusionsWe designed a novel method of root tracking that accurately and automatically measures both static and dynamic RSA parameters from a novel high-throughput root phenotyping platform. It has been used to characterize developing patterns of root systems grown under various environmental conditions. It provides a solid basis to explore the GxE interaction controlling the dynamics of root system architecture adaptive responses. In future work, our approach will be adapted to a wider range of imaging configurations and species.
Published: 2022

30. Large-scale Knowledge Distillation with Elastic Heterogeneous Computing Resources

Author: Ji Liu, Daxiang Dong, Xi Wang, An Qin, Xingjian Li, Patrick Valduriez, Dejing Dou, Dianhai Yu, Baidu Research, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)
Subjects: FOS: Computer and information sciences, Computer Science - Machine Learning, Computer Networks and Communications, Computer Science - Artificial Intelligence, Knowledge distillation, Deep neural network, Distributed computing, Computer Science Applications, Theoretical Computer Science, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Machine Learning (cs.LG), Artificial Intelligence (cs.AI), Computational Theory and Mathematics, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Computer Science - Distributed, Parallel, and Cluster Computing, Distributed, Parallel, and Cluster Computing (cs.DC), [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], Software
Abstract: Although more layers and more parameters generally improve the accuracy of the models, such big models generally have high computational complexity and require big memory, which exceed the capacity of small devices for inference and incurs long training time. In addition, it is difficult to afford long training time and inference time of big models even in high performance servers, as well. As an efficient approach to compress a large deep model (a teacher model) to a compact model (a student model), knowledge distillation emerges as a promising approach to deal with the big models. Existing knowledge distillation methods cannot exploit the elastic available computing resources and correspond to low efficiency. In this paper, we propose an Elastic Deep Learning framework for knowledge Distillation, i.e., EDL-Dist. The advantages of EDL-Dist are three-fold. First, the inference and the training process is separated. Second, elastic available computing resources can be utilized to improve the efficiency. Third, fault-tolerance of the training and inference processes is supported. We take extensive experimentation to show that the throughput of EDL-Dist is up to 3.125 times faster than the baseline method (online knowledge distillation) while the accuracy is similar or higher., To appear in Concurrency and Computation: Practice and Experience, 16 pages, 7 figures, 5 tables
Published: 2022

31. A root functional-structural model allows to assess effects of water deficit on water and solute transport parameters

Author: Bauget, Fabrice, Protto, Virginia, Pradal, Christophe, Boursiac, Yann, Maurel, Christophe, Institut des Sciences des Plantes de Montpellier (IPSIM), Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro Montpellier, Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Université de Montpellier (UM), Amélioration génétique et adaptation des plantes méditerranéennes et tropicales (UMR AGAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro Montpellier, Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)
Subjects: root system architecture, water transport, model, solute transport, xylem, root hydraulics, [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation, conductance, water deficit
Abstract: International audience; Root water uptake is driven by a combination of hydrostatic and osmotic forces. Water transport was characterized in primary roots of maize seedlings grown hydroponically under standard and water deficit (WD) conditions, as induced by addition of 150 g.L -1 polyethylene glycol-8000 (water potential= -0.336MPa). Flow measurements were performed by the pressure chamber technique in intact roots or on progressively cut root system architectures (RSA). To account for the concomitant transport of water and solutes in roots under WD, we developed within realistic RSAs a Hydraulic Tree Model integrating both solute pumping and leak. This model explains the high spontaneous sap exudation of roots grown in standard conditions, the non-linearity of pressure-to flow relationships, and negative fluxes observed under WD conditions at low external hydrostatic pressure. The model also reveals the heterogeneity of driving forces and elementary radial flows throughout RSA, and how this heterogeneity depends on both plant treatment and water transport mode. The full set of flow measurement data obtained in individual roots grown under standard or WD conditions was used in an inverse modeling approach to determine their respective radial and axial hydraulic conductivities. This approach allows to resolve dramatic effects of WD on these two components.
Published: 2022

32. Towards smart and sustainable development of modern berry cultivars in Europe

Author: Elisa Senger, Sonia Osorio, Klaus Olbricht, Paul Shaw, Béatrice Denoyes, Jahn Davik, Stefano Predieri, Saila Karhu, Sebastian Raubach, Nico Lippi, Monika Höfer, Helen Cockerton, Christophe Pradal, Ebru Kafkas, Suzanne Litthauer, Iraida Amaya, Björn Usadel, Bruno Mezzetti, Forschungszentrum Jülich GmbH | Centre de recherche de Juliers, Helmholtz-Gemeinschaft = Helmholtz Association, Instituto de Hortofruticultura Subtropical y Mediterranea 'La Mayora' (IHSM), Universidad de Málaga [Málaga] = University of Málaga [Málaga]-Consejo Superior de Investigaciones Científicas [Madrid] (CSIC), Hansabred [Dresden], The James Hutton Institute, Biologie du fruit et pathologie (BFP), Université de Bordeaux (UB)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), Norsk institutt for bioøkonomi=Norwegian Institute of Bioeconomy Research (NIBIO), Institute for BioEconomy [Sesto Fiorentino] (IBE | CNR), National Research Council of Italy | Consiglio Nazionale delle Ricerche (CNR), Natural Resources Institute Finland (LUKE), Julius Kühn-Institut - Federal Research Centre for Cultivated Plants (JKI), National Institute of Agricultural Botany (NIAB), Amélioration génétique et adaptation des plantes méditerranéennes et tropicales (UMR AGAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro Montpellier, Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Université de Montpellier (UM), Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Cukurova University, Heinrich Heine Universität Düsseldorf = Heinrich Heine University [Düsseldorf], Università Politecnica delle Marche [Ancona] (UNIVPM), and The authors acknowledge support from the European Union’s Horizon 2020 research and innovation program (grant agreement ID: 101000747)
Subjects: consumer preference, [SDV]Life Sciences [q-bio], métadonnées, plant genetic resources, Plant Science, Fragaria, F30 - Génétique et amélioration des plantes, génomique, Ressource génétique végétale, image analysis, Genetics, genomics, [SDV.BV]Life Sciences [q-bio]/Vegetal Biology, Humans, amélioration génétique, QK, BreedingValue project, trait/genotype association, Cell Biology, Sustainable Development, metabolomics, Amélioration des plantes, Petits fruits, Propriété organoleptique, [SDV.BV.AP]Life Sciences [q-bio]/Vegetal Biology/Plant breeding, Valeur génétique, Plant Breeding, ddc:580, berry breeding, Fruit, Rubus, Vaccinium, Génotype, Genome-Wide Association Study
Abstract: International audience; Fresh berries are a popular and important component of the human diet. The demand for high-quality berries and sustainable production methods is increasing globally, challenging breeders to develop modern berry cultivars that fulfill all desired characteristics. Since 1994, research projects have characterized genetic resources, developed modern tools for high-throughput screening, and published data in publicly available repositories. However, the key findings of different disciplines are rarely linked together and only a limited range of traits and genotypes has been investigated. The Horizon2020 project BreedingValue will address these challenges by studying a broader panel of strawberry, raspberry and blueberry genotypes in detail, in order to recover the lost genetic diversity that has limited the aroma and flavor intensity of recent cultivars. We will combine metabolic analysis with sensory panel tests and surveys to identify the key components of taste, flavor and aroma in berries across Europe, leading to a high-resolution map of quality requirements for future berry cultivars. Traits linked to berry yields and the effect of environmental stress will be investigated using modern image analysis methods and modeling. We will also use genetic analysis to determine the genetic basis of complex traits for the development and optimization of modern breeding technologies such as molecular marker arrays, genomic selection and genome wide association studies. Finally, the results, raw data and metadata will be made publicly available on the open platform Germinate in order to meet FAIR data principles and provide the basis for sustainable research in the future.
Published: 2022

33. Very High Resolution Species Distribution Modeling Based on Remote Sensing Imagery: How to Capture Fine-Grained and Large-Scale Vegetation Ecology With Convolutional Neural Networks?

Author: Benjamin Deneu, Alexis Joly, Pierre Bonnet, Maximilien Servajean, François Munoz, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Université Paul-Valéry - Montpellier 3 (UPVM), ADVanced Analytics for data SciencE (ADVANSE), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Laboratoire Interdisciplinaire de Physique [Saint Martin d’Hères] (LIPhy ), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA), and ANR-16-CONV-0004,DIGITAG,Institut Convergences en Agriculture Numérique(2016)
Subjects: [SDV.EE.ECO]Life Sciences [q-bio]/Ecology, environment/Ecosystems, Remote-sensing imagery, Plant functional traits, Species distribution model, Convolutional neural network, Plant Science, [SDE.BE]Environmental Sciences/Biodiversity and Ecology, [SDV.BV.BOT]Life Sciences [q-bio]/Vegetal Biology/Botanics, Ecological interpretation, [SDV.BID.SPT]Life Sciences [q-bio]/Biodiversity/Systematics, Phylogenetics and taxonomy, t-SNE, Very fine scale prediction
Abstract: Species Distribution Models (SDMs) are fundamental tools in ecology for predicting the geographic distribution of species based on environmental data. They are also very useful from an application point of view, whether for the implementation of conservation plans for threatened species or for monitoring invasive species. The generalizability and spatial accuracy of an SDM depend very strongly on the type of model used and the environmental data used as explanatory variables. In this article, we study a country-wide species distribution model based on very high resolution (VHR) (1 m) remote sensing images processed by a convolutional neural network. We demonstrate that this model can capture landscape and habitat information at very fine spatial scales while providing overall better predictive performance than conventional models. Moreover, to demonstrate the ecological significance of the model, we propose an original analysis based on the t-distributed Stochastic Neighbor Embedding (t-SNE) dimension reduction technique. It allows visualizing the relation between input data and species traits or environment learned by the model as well as conducting some statistical tests verifying them. We also analyze the spatial mapping of the t-SNE dimensions at both national and local levels, showing the model benefit of automatically learning environmental variation at multiple scales.
Published: 2022

34. Variable size segmentation for efficient representation and querying of non-uniform time series datasets

Author: Djebour, Lamia, Akbarinia, Reza, Masseglia, Florent, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)
Subjects: Nearest-neighbor search Time Series, Representations, Time series, [INFO.INFO-IR]Computer Science [cs]/Information Retrieval [cs.IR], Spatial-temporal systems, Information Retrieval, Nearest-neighbor search, Data mining
Abstract: International audience; Existing approaches for time series similarity computing are the core of many data analytics tasks. Given the considered data volumes, or simply the need for fast response times, they often rely on shorter representations, usually with information loss. This incurs approximate comparisons where precision is a major issue. We present and experimentally evaluate ASAX, a new approach for segmenting time series before their transformation into symbolic representations. ASAX reduces significantly the information loss incurred by possible splittings at different steps of the representation calculation, particularly for datasets with unbalanced (nonuniform) distributions. We provide theoretical guarantees on the lower bound of similarity measures, and our experiments illustrate that our method outperforms the state of the art, with significant gain in precision for datasets with unbalanced distributions.
Published: 2022

35. LifeCLEF 2022 Teaser: An Evaluation of Machine-Learning Based Species Identification and Species Distribution Prediction

Author: Alexis Joly, Hervé Goëau, Stefan Kahl, Lukáš Picek, Titouan Lorieul, Elijah Cole, Benjamin Deneu, Maximilien Servajean, Andrew Durso, Isabelle Bolon, Hervé Glotin, Robert Planqué, Willem-Pier Vellinga, Holger Klinck, Tom Denton, Ivan Eggel, Pierre Bonnet, Henning Müller, Milan Šulc, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Cornell University [New York], University of West Bohemia [Plzeň ], Computer Science Department (CS CALTECH), California Institute of Technology (CALTECH), ADVanced Analytics for data SciencE (ADVANSE), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Florida Gulf Coast University, Université de Genève = University of Geneva (UNIGE), Laboratoire d'Informatique et Systèmes (LIS), Aix Marseille Université (AMU)-Université de Toulon (UTLN)-Centre National de la Recherche Scientifique (CNRS), Xeno-canto foundation, Google LLC, Haute Ecole Spécialisée de Suisse Occidentale (HES-SO), Czech Technical University in Prague (CTU), DigitAG, Matthias Hagen, Suzan Verberne, Craig Macdonald, Christin Seifert, Krisztian Balog, Kjetil Nørvåg, Vinay Setty, and European Project: 863463,Cos4Cloud project
Subjects: Bird identification, Machine-Learning, Species distribution model, Biodiversity, [SDV.BV.BOT]Life Sciences [q-bio]/Vegetal Biology/Botanics, [SDV.BID.SPT]Life Sciences [q-bio]/Biodiversity/Systematics, Phylogenetics and taxonomy, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], [SDV.EE.ECO]Life Sciences [q-bio]/Ecology, environment/Ecosystems, AI, Machine learning, Species Identification, [INFO]Computer Science [cs], LifeCLEF, Plant identification, [SDE.BE]Environmental Sciences/Biodiversity and Ecology, Species prediction, Snake identification
Abstract: International audience; Building accurate knowledge of the identity, the geographic distribution and the evolution of species is essential for the sustainable development of humanity, as well as for biodiversity conservation. However, the difficulty of identifying plants, animals and fungi is hindering the aggregation of new data and knowledge. Identifying and naming living organisms is almost impossible for the general public and is often difficult even for professionals and naturalists. Bridging this gap is a key step towards enabling effective biodiversity monitoring systems. The LifeCLEF campaign, presented in this paper, has been promoting and evaluating advances in this domain since 2011. The 2022 edition proposes five data-oriented challenges related to the identification and prediction of biodiversity: (i) PlantCLEF: very large-scale plant identification, (ii) BirdCLEF: bird species recognition in audio soundscapes, (iii) GeoLifeCLEF: remote sensing based prediction of species, (iv) SnakeCLEF: Snake Species Identification in Medically Important scenarios, and (v) FungiCLEF: Fungi recognition from images and metadata.
Published: 2022

36. Deep Species Distribution Modeling From Sentinel-2 Image Time-Series: A Global Scale Analysis on the Orchid Family

Author: Joaquim Estopinan, Maximilien Servajean, Pierre Bonnet, François Munoz, Alexis Joly, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), ADVanced Analytics for data SciencE (ADVANSE), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Laboratoire Interdisciplinaire de Physique [Saint Martin d’Hères] (LIPhy ), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA), and CACTUS Fund - INRIA
Subjects: [SDV.EE.ECO]Life Sciences [q-bio]/Ecology, environment/Ecosystems, Deep learning, Convolutional neural networks, Plant Science, Remote sensing, Sentinel-2, [SDV.BV.BOT]Life Sciences [q-bio]/Vegetal Biology/Botanics, [SDE.BE]Environmental Sciences/Biodiversity and Ecology, Macroecology, Image time-series, [SDV.BID.SPT]Life Sciences [q-bio]/Biodiversity/Systematics, Phylogenetics and taxonomy, Data science, Species distribution modeling
Abstract: Species distribution models (SDMs) are widely used numerical tools that rely on correlations between geolocated presences (and possibly absences) and environmental predictors to model the ecological preferences of species. Recently, SDMs exploiting deep learning and remote sensing images have emerged and have demonstrated high predictive performance. In particular, it has been shown that one of the key advantages of these models (called deep-SDMs) is their ability to capture the spatial structure of the landscape, unlike prior models. In this paper, we examine whether the temporal dimension of remote sensing images can also be exploited by deep-SDMs. Indeed, satellites such as Sentinel-2 are now providing data with a high temporal revisit, and it is likely that the resulting time-series of images contain relevant information about the seasonal variations of the environment and vegetation. To confirm this hypothesis, we built a substantial and original dataset (calledDeepOrchidSeries) aimed at modeling the distribution of orchids on a global scale based on Sentinel-2 image time series. It includes around 1 million occurrences of orchids worldwide, each being paired with a 12-month-long time series of high-resolution images (640 x 640 m RGB+IR patches centered on the geolocated observations). This ambitious dataset enabled us to train several deep-SDMs based on convolutional neural networks (CNNs) whose input was extended to include the temporal dimension. To quantify the contribution of the temporal dimension, we designed a novel interpretability methodology based on temporal permutation tests, temporal sampling, and temporal averaging. We show that the predictive performance of the model is greatly increased by the seasonality information contained in the temporal series. In particular, occurrence-poor species and diversity-rich regions are the ones that benefit the most from this improvement, revealing the importance of habitat's temporal dynamics to characterize species distribution.
Published: 2022

37. How Networks of Citizen Observatories Can Increase the Quality and Quantity of Citizen-Science-Generated Data Used to Monitor SDG Indicators

Author: Sasha Marie Woods, Maria Daskolia, Alexis Joly, Pierre Bonnet, Karen Soacha, Sonia Liñan, Tim Woods, Jaume Piera, Luigi Ceccaroni, Earthwatch Institute, National and Kapodistrian University of Athens (NKUA), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Spanish National Research Council (CSIC), Open University of Catalonia [Barcelona], ICM-CSIC authors acknowledge the institutional support of the 'Severo Ochoa Centre of Excellence' accreditation (CEX2019-000928-S, European Project: 863463,Cos4Cloud project, European Commission, and Agencia Estatal de Investigación (España)
Subjects: Renewable Energy, Sustainability and the Environment, Geography, Planning and Development, Sustainable Development Goals, citizen science, citizen observatory, interoperability, co-design, education, Management, Monitoring, Policy and Law, Citizen science, [SDV.BV.BOT]Life Sciences [q-bio]/Vegetal Biology/Botanics, Interoperability, [SDV.BID.SPT]Life Sciences [q-bio]/Biodiversity/Systematics, Phylogenetics and taxonomy, Citizen observatory, Education, [SDV.EE.ECO]Life Sciences [q-bio]/Ecology, environment/Ecosystems, ComputerSystemsOrganization_MISCELLANEOUS, Co-design, [SDE.BE]Environmental Sciences/Biodiversity and Ecology
Abstract: Special Issue Citizen Science Projects for Environmental Challenges and Sustainable Development Goals.-- 17 pages, 2 pages, 1 figure.-- Data Availability Statement: Not applicable, There is a growing acknowledgement that citizen observatories, and other forms of citizen-generated data, have a significant role in tracking progress towards the Sustainable Development Goals. This is evident in the increasing number of Sustainable Development Goals’ indicators for which such data are already being used and in the high-level recognition of the potential role that citizen science can play. In this article, we argue that networks of citizen observatories will help realise this potential. Drawing on the Cos4Cloud project as an example, we highlight how such networks can make citizen-generated data more interoperable and accessible (among other qualities), increasing their impact and usefulness. Furthermore, we highlight other, perhaps overlooked, advantages of citizen observatories and citizen-generated data: educating and informing citizen scientists about the Sustainable Development Goals and co-creating solutions to the global challenges they address, The research described in this paper was funded by the European Commission via the Cos4Cloud and MICS projects, which have received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreements 863463 and 824711. ICM-CSIC authors acknowledge the institutional support of the “Severo Ochoa Centre of Excellence” accreditation (CEX2019-000928-S)
Published: 2022

38. Making the Right Move to Senior Researcher

Author: Patrick Valduriez, Scientific Data Management (ZENITH), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), and Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Inria Sophia Antipolis - Méditerranée (CRISAM)
Subjects: [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], Point (typography), business.industry, Computer science, Data management, Nice, 0102 computer and information sciences, 02 engineering and technology, Public relations, 01 natural sciences, Work (electrical), 010201 computation theory & mathematics, 020204 information systems, Institution (computer science), Research environment, 0202 electrical engineering, electronic engineering, information engineering, business, computer, Software, Information Systems, computer.programming_language
Abstract: International audience; I have been working on research in data management for the last 40 years. I like my job and my research institution (Inria, the French national research institute for computer science), which have offered me great opportunities to learn a lot, do good work, get to know smart and nice people and overall feel useful. However, since the early days of my mid-career, the research environment, including academia and industry, has certainly become more complex, making the move from junior (or pre-tenure) researcher to senior researcher quite challenging. Based on my experience, I review some of the main questions and challenges and give some hints on how to deal with them. I'll sometimes use stories and anecdotes to illustrate the point.
Published: 2021

39. BestNeighbor: efficient evaluation of kNN queries on large time series databases

Author: Patrick Valduriez, Djamel Edine Yagoubi, Themis Palpanas, Boyan Kolev, Florent Masseglia, Reza Akbarinia, Oleksandra Levchenko, Dennis Shasha, Scientific Data Management (ZENITH), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), LeanXcale [Madrid], Université de Paris (UP), New York University [New York] (NYU), NYU System (NYU), Université Paris Cité (UPCité), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Inria Sophia Antipolis - Méditerranée (CRISAM)
Subjects: Distributed Querying, Computer science, Random projection, Time Series, 02 engineering and technology, computer.software_genre, Synthetic data, k-nearest neighbors algorithm, Set (abstract data type), Artificial Intelligence, 020204 information systems, Spark (mathematics), 0202 electrical engineering, electronic engineering, information engineering, Pruning (decision trees), Fourier Coefficients, Fourier series, Database, Series (mathematics), Frequency Analysis, Human-Computer Interaction, Parallel Indexing, Hardware and Architecture, [INFO.INFO-IR]Computer Science [cs]/Information Retrieval [cs.IR], 020201 artificial intelligence & image processing, Filtering, computer, Software, Information Systems
Abstract: International audience; This paper presents parallel solutions (developed based on two state-of-the-art algorithms iSAX and sketch) for evaluating kNN (k nearest neighbor) queries on large databases of time series, compares them based on various measures of quality and time performance, offers a tool that uses the characteristics of application data to determine which algorithm to choose for that application and how to set the parameters for that algorithm. Specifically, our experiments show that: (i) iSAX and its derivatives perform best in both time and quality when the time series can be characterized by a few low frequency Fourier Coefficients , a regime where the iSAX pruning approach works well. (ii) iSAX performs significantly less well when high frequency Fourier Coefficients have much of the energy of the time series. (iii) A random projection approach based on sketches by contrast is more or less independent of the frequency power spectrum. The experiments show the close relationship between pruning ratio and time for exact iSAX as well as between pruning ratio and the quality of approximate iSAX. Our toolkit analyzes typical time series of an application (i) to determine optimal segment sizes for iSAX and (ii) when to use Parallel Sketches instead of iSAX. Our algorithms have been implemented using Spark, evaluated over a cluster of nodes, and have been applied to both real and synthetic data. The results apply to any databases of numerical sequences, whether or not they relate to time.
Published: 2020

40. Simulating the effects of water limitation on plant biomass using a 3D functional-structural plant model of shoot and root driven by soil hydraulics

Author: Renato K. Braghiere, Loïc Pagès, Frédéric Gérard, Christophe Pradal, Jochem B. Evers, Ecologie fonctionnelle et biogéochimie des sols et des agro-écosystèmes (UMR Eco&Sols), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut de Recherche pour le Développement (IRD)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro - Montpellier SupAgro, Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro), California Institute of Technology (CALTECH), University of California [Los Angeles] (UCLA), University of California (UC), Wageningen University and Research [Wageningen] (WUR), Amélioration génétique et adaptation des plantes méditerranéennes et tropicales (UMR AGAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro - Montpellier SupAgro, Scientific Data Management (ZENITH), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Unité de recherche Plantes et Systèmes de Culture Horticoles (PSH), Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), contract with the National Aeronautics and Space Administration (80NM0018D0004)., European Project: 727217,ReMIX(2017), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut de Recherche pour le Développement (IRD)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro), Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), University of California, Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Inria Sophia Antipolis - Méditerranée (CRISAM)
Subjects: 0106 biological sciences, 010504 meteorology & atmospheric sciences, Biomass, GroIMP, F62 - Physiologie végétale - Croissance et développement, Déficit hydrique du sol, Plant Science, Root system, Plant Roots, 01 natural sciences, Soil, santé des plantes, Water content, media_common, Transpiration, water deficit, 2. Zero hunger, U10 - Informatique, mathématiques et statistiques, ArchiSimple, Propriété hydraulique du sol, PE&RC, soil-plant interactions, Physiologie végétale, Droughts, Crop and Weed Ecology, Plant Shoots, water uptake, Stomatal conductance, Relation plante sol, media_common.quotation_subject, Biology, Competition (biology), Culture intercalaire, functional-structural plant models, Min3P, [SDU.ENVI]Sciences of the Universe [physics]/Continental interfaces, environment, Absorption d'eau, 0105 earth and related environmental sciences, photosynthesis, Water, Original Articles, Interspecific competition, 15. Life on land, [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation, Plant Leaves, Agronomy, Modélisation, Soil water, soil modelling, intercropping, 010606 plant biology & botany
Abstract: Background and AimsImproved modelling of carbon assimilation and plant growth to low soil moisture requires evaluation of underlying mechanisms in the soil, roots, and shoots. The feedback between plants and their local environment throughout the whole spectrum soil-root-shoot-environment is crucial to accurately describe and evaluate the impact of environmental changes on plant development. This study presents a 3D functional structural plant model, in which shoot and root growth are driven by radiative transfer, photosynthesis, and soil hydrodynamics through different parameterisation schemes relating soil water deficit and carbon assimilation. The new coupled model is used to evaluate the impact of soil moisture availability on plant productivity for two different groups of flowering plants under different spatial configurations.MethodsIn order to address different aspects of plant development due to limited soil water availability, a 3D FSP model including root, shoot, and soil was constructed by linking three different well-stablished models of airborne plant, root architecture, and reactive transport in the soil. Different parameterisation schemes were used in order to integrate photosynthetic rate with root water uptake within the coupled model. The behaviour of the model was assessed on how the growth of two different types of plants, i.e. monocot and dicot, is impacted by soil water deficit under different competitive conditions: isolated (no competition), intra, and interspecific competition.Key ResultsThe model proved to be capable of simulating carbon assimilation and plant development under different growing settings including isolated monocots and dicots, intra, and interspecific competition. The model predicted that (1) soil water availability has a larger impact on photosynthesis than on carbon allocation; (2) soil water deficit has an impact on root and shoot biomass production by up to 90 % for monocots and 50 % for dicots; and (3) the improved dicot biomass production in interspecific competition was highly related to root depth and plant transpiration.ConclusionsAn integrated model of 3D shoot architecture and biomass development with a 3D root system representation, including light limitation and water uptake considering soil hydraulics, was presented. Plant-plant competition and regulation on stomatal conductance to drought were able to be predicted by the model. In the cases evaluated here, water limitation impacted plant growth almost 10 times more than the light environment.
Published: 2020

41. Can Artificial Intelligence Help in the Study of Vegetative Growth Patterns from Herbarium Collections? An Evaluation of the Tropical Flora of the French Guiana Forest

Author: Hervé Goëau, Titouan Lorieul, Patrick Heuret, Alexis Joly, Pierre Bonnet, Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), ANR-10-LABX-0025,CEBA,CEnter of the study of Biodiversity in Amazonia(2010), and ANR-11-INBS-0004,E-RECOLNAT,Valorisation de 350 ans de collections d'histoire naturelle : une plateforme numérique(2011)
Subjects: Ecology, Deep learning, Plant Science, [SDV.BV.BOT]Life Sciences [q-bio]/Vegetal Biology/Botanics, [SDV.BID.SPT]Life Sciences [q-bio]/Biodiversity/Systematics, Phylogenetics and taxonomy, Neural network, Natural history collection, [SDV.EE.ECO]Life Sciences [q-bio]/Ecology, environment/Ecosystems, Machine learning, Visual analysis, Vegetative phenology, [SDE.BE]Environmental Sciences/Biodiversity and Ecology, vegetative phenology, plant growth, herbarium specimens, natural history collection, deep learning, machine learning, neural network, visual analysis, Ecology, Evolution, Behavior and Systematics, Plant growth, Herbarium specimens
Abstract: International audience; A better knowledge of tree vegetative growth phenology and its relationship to environmental variables is crucial to understanding forest growth dynamics and how climate change may affect it. Less studied than reproductive structures, vegetative growth phenology focuses primarily on the analysis of growing shoots, from buds to leaf fall. In temperate regions, low winter temperatures impose a cessation of vegetative growth shoots and lead to a well-known annual growth cycle pattern for most species. The humid tropics, on the other hand, have less seasonality and contain many more tree species, leading to a diversity of patterns that is still poorly known and understood. The work in this study aims to advance knowledge in this area, focusing specifically on herbarium scans, as herbariums offer the promise of tracking phenology over long periods of time. However, such a study requires a large number of shoots to be able to draw statistically relevant conclusions. We propose to investigate the extent to which the use of deep learning can help detect and type-classify these relatively rare vegetative structures in herbarium collections. Our results demonstrate the relevance of using herbarium data in vegetative phenology research as well as the potential of deep learning approaches for growing shoot detection.
Published: 2022
Full Text: View/download PDF

42. Principled methods for mixtures processing

Author: Liutkus, Antoine, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Université de Montpellier, and Christian Jutten(christian.jutten@gipsa-lab.grenoble-inp.fr)
Subjects: machine learning, séparation de sources, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], source separation, audio, [INFO]Computer Science [cs], Stochastic processes & statistics
Abstract: This document is my thesis for getting the habilitation à diriger des recherches, which is the french diploma that is required to fully supervise Ph.D. students. It summarizes the research I did in the last 15 years and also provides the shortterm research directions and applications I want to investigate. Regarding my past research, I first describe the work I did on probabilistic audio modeling, including the separation of Gaussian and αstable stochastic processes. Then, I mention my work on deep learning applied to audio, which rapidly turned into a large effort for community service. Finally, I present my contributions in machine learning, with some works on hardware compressed sensing and probabilistic generative models.My research programme involves a theoretical part that revolves around probabilistic machine learning, and an applied part that concerns the processing of time series arising in both audio and life sciences.
Published: 2022

43. Differentially Private Coordinate Descent for Composite Empirical Risk Minimization

Author: Mangold, Paul, Bellet, Aurélien, Salmon, Joseph, Tommasi, Marc, Machine Learning in Information Networks (MAGNET), Inria Lille - Nord Europe, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), Institut Montpelliérain Alexander Grothendieck (IMAG), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Université de Montpellier (UM), Institut Universitaire de France (IUF), Ministère de l'Education nationale, de l’Enseignement supérieur et de la Recherche (M.E.N.E.S.R.), This work was supported in part by the Inria Exploratory Action FLAMED, ANR-20-CE23-0015,PRIDE,Apprentissage automatique décentralisé et préservant la vie privée(2020), ANR-20-CHIA-0001,CAMELOT,Apprentissage automatique et optimisation coopératifs.(2020), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS), Centre National de la Recherche Scientifique (CNRS), Université de Lille-Centrale Lille-Centre National de la Recherche Scientifique (CNRS)-Université de Lille-Centrale Lille-Centre National de la Recherche Scientifique (CNRS), Machine Learning in Information Networks [MAGNET], Institut Montpelliérain Alexander Grothendieck [IMAG], and Université de Montpellier [UM]
Subjects: FOS: Computer and information sciences, Computer Science - Machine Learning, Computer Science - Cryptography and Security, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], Statistics - Machine Learning, Machine Learning (stat.ML), Cryptography and Security (cs.CR), Machine Learning (cs.LG)
Abstract: Machine learning models can leak information about the data used to train them. To mitigate this issue, Differentially Private (DP) variants of optimization algorithms like Stochastic Gradient Descent (DP-SGD) have been designed to trade-off utility for privacy in Empirical Risk Minimization (ERM) problems. In this paper, we propose Differentially Private proximal Coordinate Descent (DP-CD), a new method to solve composite DP-ERM problems. We derive utility guarantees through a novel theoretical analysis of inexact coordinate descent. Our results show that, thanks to larger step sizes, DP-CD can exploit imbalance in gradient coordinates to outperform DP-SGD. We also prove new lower bounds for composite DP-ERM under coordinate-wise regularity assumptions, that are nearly matched by DP-CD. For practical implementations, we propose to clip gradients using coordinate-wise thresholds that emerge from our theory, avoiding costly hyperparameter tuning. Experiments on real and synthetic data support our results, and show that DP-CD compares favorably with DP-SGD., 36 pages, 3 figures
Published: 2022

44. How does the degree of novelty impacts semi-supervised representation learning for novel class retrieval?

Author: Leroy, Quentin, Buisson, Olivier, Joly, Alexis, Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), and Institut National de l'Audiovisuel (INA)
Subjects: FOS: Computer and information sciences, Artificial Intelligence (cs.AI), [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Computer Science - Artificial Intelligence, Computer Vision and Pattern Recognition (cs.CV), [INFO.INFO-MM]Computer Science [cs]/Multimedia [cs.MM], Computer Science - Computer Vision and Pattern Recognition, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], [SDE.BE]Environmental Sciences/Biodiversity and Ecology, [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation, [SDE.ES]Environmental Sciences/Environmental and Society, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract: Supervised representation learning with deep networks tends to overfit the training classes and the generalization to novel classes is a challenging question. It is common to evaluate a learned embedding on held-out images of the same training classes. In real applications however, data comes from new sources and novel classes are likely to arise. We hypothesize that incorporating unlabelled images of novel classes in the training set in a semi-supervised fashion would be beneficial for the efficient retrieval of novel-class images compared to a vanilla supervised representation. To verify this hypothesis in a comprehensive way, we propose an original evaluation methodology that varies the degree of novelty of novel classes by partitioning the dataset category-wise either randomly, or semantically, i.e. by minimizing the shared semantics between base and novel classes. This evaluation procedure allows to train a representation blindly to any novel-class labels and evaluate the frozen representation on the retrieval of base or novel classes. We find that a vanilla supervised representation falls short on the retrieval of novel classes even more so when the semantics gap is higher. Semi-supervised algorithms allow to partially bridge this performance gap but there is still much room for improvement.
Published: 2022
Full Text: View/download PDF

45. PhenoTrack3D: An automatic high-throughput phenotyping pipeline to track maize organs over time

Author: Benoit Daviet, Romain Fernandez, Llorenç Cabrera-Bosquet, Christophe Pradal, Christian Fournier, Écophysiologie des Plantes sous Stress environnementaux (LEPSE), Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro Montpellier, Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro), Amélioration génétique et adaptation des plantes méditerranéennes et tropicales (UMR AGAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro Montpellier, Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Université de Montpellier (UM), Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), European Project: 731013 ,EPPN2020(2017), and European Project: 314061,EC:FP7:TPT,FP7-AAT-2012-RTD-1,STARGATE(2012)
Subjects: Sequence alignment, Tracking, Plant physiology, Genetics, High-throughput phenotyping, Openalea, Computer vision, Plant Science, [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation, Biotechnology, Maize
Abstract: Background High-throughput phenotyping platforms allow the study of the form and function of a large number of genotypes subjected to different growing conditions (GxE). A number of image acquisition and processing pipelines have been developed to automate this process, for micro-plots in the field and for individual plants in controlled conditions. Capturing shoot development requires extracting from images both the evolution of the 3D plant architecture as a whole, and a temporal tracking of the growth of its organs. Results We propose PhenoTrack3D, a new pipeline to extract a 3D + t reconstruction of maize. It allows the study of plant architecture and individual organ development over time during the entire growth cycle. The method tracks the development of each organ from a time-series of plants whose organs have already been segmented in 3D using existing methods, such as Phenomenal [Artzet et al. in BioRxiv 1:805739, 2019] which was chosen in this study. First, a novel stem detection method based on deep-learning is used to locate precisely the point of separation between ligulated and growing leaves. Second, a new and original multiple sequence alignment algorithm has been developed to perform the temporal tracking of ligulated leaves, which have a consistent geometry over time and an unambiguous topological position. Finally, growing leaves are back-tracked with a distance-based approach. This pipeline is validated on a challenging dataset of 60 maize hybrids imaged daily from emergence to maturity in the PhenoArch platform (ca. 250,000 images). Stem tip was precisely detected over time (RMSE Conclusions We developed a novel phenotyping method based on sequence alignment and deep-learning. It allows to characterise the development of maize architecture at organ level, automatically and at a high-throughput. It has been validated on hundreds of plants during the entire development cycle, showing its applicability on GxE analyses of large maize datasets.
Published: 2022

46. Dynamic Species Distribution Modeling Reveals the Pivotal Role of Human-Mediated Long-Distance Dispersal in Plant Invasion

Author: Botella, Christophe, Bonnet, Pierre, Hui, Cang, Joly, Alexis, Richardson, David, Stellenbosch University, Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)
Subjects: Bayesian inference, citizen science, population dynamics, long-distance dispersal, matrix population models, presence-only data, [SDE.BE]Environmental Sciences/Biodiversity and Ecology, [SDV.BV.BOT]Life Sciences [q-bio]/Vegetal Biology/Botanics, species distribution models, human-mediated dispersal
Abstract: International audience; Plant invasions generate massive ecological and economic costs worldwide. Predicting their spatial dynamics is crucial to the design of effective management strategies and the prevention of invasions. Earlier studies highlighted the crucial role of long-distance dispersal in explaining the speed of many invasions. In addition, invasion speed depends highly on the duration of its lag phase, which may depend on the scaling of fecundity with age, especially for woody plants, even though empirical proof is still rare. Bayesian dynamic species distribution models enable the fitting of process-based models to partial and heterogeneous observations using a state-space modeling approach, thus offering a tool to test such hypotheses on past invasions over large spatial scales. We use such a model to explore the roles of long-distance dispersal and age-structured fecundity in the transient invasion dynamics of Plectranthus barbatus, a woody plant invader in South Africa. Our lattice-based model accounts for both short and human-mediated long-distance dispersal, as well as age-structured fecundity. We fitted our model on opportunistic occurrences, accounting for the spatio-temporal variations of the sampling effort and the variable detection rates across datasets. The Bayesian framework enables us to integrate a priori knowledge on demographic parameters and control identifiability issues. The model revealed a massive wave of spatial spread driven by human-mediated long-distance dispersal during the first decade and a subsequent drastic population growth, leading to a global equilibrium in the mid-1990s. Without long-distance dispersal, the maximum population would have been equivalent to 30% of the current equilibrium population. We further identified the reproductive maturity at three years old, which contributed to the lag phase before the final wave of population growth. Our results highlighted the importance of the early eradication of weedy horticultural alien plants around urban areas to hamper and delay the invasive spread
Published: 2022

47. Envisaging a global infrastructure to exploit the potential of digitised collections

Author: Quentin Groom, Mathias Dillen, Wouter Addink, Arturo H. Ariño, Christian Bölling, Pierre Bonnet, Lorenzo Cecchi, Elizabeth R. Ellwood, Rui Figueira, Pierre-Yves Gagnier, Olwen Grace, Anton Güntsch, Helen Hardy, Pieter Huybrechts, Roger Hyam, Alexis Joly, Isabel Larridon, Vamsi Krishna Kommineni, Laurence Livermore, Ricardo Jorge Lopes, Jeremy Miller, Sofie Meeus, Kenzo Milleville, Marc Pignal, Renato Panda, Jorrit H. Poelen, Blagoj Ristevski, Tim Robertson, Cristina Rufino, Joaquim Santos, Maarten Schermer, Katja Seltmann, Ben Scott, Heliana Teixeira, Maarten Trekels, Jitendra Gaikwad, Meise Botanic Garden [Belgium] (Plantentuin), Naturalis Biodiversity Center [Leiden], Museum für Naturkunde [Berlin], Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM), Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Museo di Storia Naturale di Firenze, Università degli Studi di Firenze = University of Florence (UniFI), Instituto Superior de Agronomia [Lisboa] (ISA), Universidade de Lisboa = University of Lisbon (ULISBOA), Muséum national d'Histoire naturelle (MNHN), Royal Botanic Garden [Edinburgh], Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Royal Botanic Gardens [Kew], Friedrich-Schiller-Universität = Friedrich Schiller University Jena [Jena, Germany], Universidade do Porto = University of Porto, Internet Technology and Data Science Lab (IDLab), Universiteit Antwerpen = University of Antwerpen [Antwerpen]-Universiteit Gent = Ghent University (UGENT), Global Biodiversity Information Facility (GBIF), and Universidade de Aveiro
Subjects: [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], [INFO.INFO-MM]Computer Science [cs]/Multimedia [cs.MM], [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], [SDE.BE]Environmental Sciences/Biodiversity and Ecology, [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation, [SDE.ES]Environmental Sciences/Environmental and Society, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract: Tens of millions of images from biological collections have become available online in the last two decades. In parallel, there has been a dramatic increase in the capabilities of image analysis technologies, especially those involving machine learning and computer vision. Whilst image analysis has become mainstream in consumer applications, it is still only used on an artisanal basis in the biological collections community, largely because the image corpora are dispersed. Yet, there is massive untapped potential for novel applications and research if the images of collection objects could be made accessible as a single corpus. In this paper, we make the case for building infrastructure that could support image analysis of collection objects. We show that such an infrastructure is entirely feasible and well worth the investment.
Published: 2022

48. Stochastic smoothing of the top-K calibrated hinge loss for deep imbalanced classification

Author: Garcin, Camille, Servajean, Maximilien, Joly, Alexis, Salmon, Joseph, Institut Montpelliérain Alexander Grothendieck (IMAG), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Université de Montpellier (UM), ADVanced Analytics for data SciencE (ADVANSE), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Université Paul-Valéry - Montpellier 3 (UPVM), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Institut Universitaire de France (IUF), Ministère de l'Education nationale, de l’Enseignement supérieur et de la Recherche (M.E.N.E.S.R.), and ANR-20-CHIA-0001,CAMELOT,Apprentissage automatique et optimisation coopératifs.(2020)
Subjects: FOS: Computer and information sciences, Computer Science - Machine Learning, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], Statistics - Machine Learning, Machine Learning (stat.ML), GeneralLiterature_MISCELLANEOUS, Machine Learning (cs.LG)
Abstract: International audience; In modern classification tasks, the number of labels is getting larger and larger, as is the size of the datasets encountered in practice. As the number of classes increases, class ambiguity and class imbalance become more and more problematic to achieve high top-1 accuracy. Meanwhile, Top-K metrics (metrics allowing K guesses) have become popular, especially for performance reporting. Yet, proposing top-K losses tailored for deep learning remains a challenge, both theoretically and practically. In this paper we introduce a stochastic top-K hinge loss inspired by recent developments on top-K calibrated losses. Our proposal is based on the smoothing of the top-K operator building on the flexible "perturbed optimizer" framework. We show that our loss function performs very well in the case of balanced datasets, while benefiting from a significantly lower computational time than the state-of-the-art top-K loss function. In addition, we propose a simple variant of our loss for the imbalanced case. Experiments on a heavy-tailed dataset show that our loss function significantly outperforms other baseline loss functions.
Published: 2022
Full Text: View/download PDF

49. TSPred: A framework for nonstationary time series prediction

Author: Eduardo Bezerra, Fabio Porto, Rebecca Salles, Eduardo Ogasawara, Esther Pacitti, Centro Federal de Educação Tecnológica Celso Suckow da Fonseca (Rio de Janeiro) ( CEFET/RJ), Scientific Data Management (ZENITH), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Laboratorio Nacional de Computação Cientifica [Rio de Janeiro] (LNCC / MCT), Associated team Hpdasc, Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)
Subjects: Computer science, Process (engineering), Cognitive Neuroscience, Data transformation (statistics), 02 engineering and technology, computer.software_genre, 01 natural sciences, 010104 statistics & probability, Artificial Intelligence, Prediction methods, 0202 electrical engineering, electronic engineering, information engineering, Decomposition (computer science), Preprocessor, 0101 mathematics, Time series, preprocessing, [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], Series (mathematics), nonstationarity, prediction, Computer Science Applications, machine learning, transform, 020201 artificial intelligence & image processing, Data mining, time series, computer
Abstract: International audience; The nonstationary time series prediction is challenging since it demands knowledge of both data transformation and prediction methods. This paper presents TSPred, a framework for nonstationary time series prediction. It differs from the mainstream frameworks since it establishes a prediction process that seamlessly integrates nonstationary time series transformations with state-of-the-art statistical and machine learning methods. It is made available as an R-package, which provides functions for defining and conducting time series prediction, including data pre(post)processing, decomposition, modeling, prediction, and accuracy assessment. Besides, TSPred enables user-defined methods, which significantly expands the applicability of the framework.
Published: 2022

50. Phenotyping and modeling of root hydraulic architecture reveal critical determinants of axial water transport

Author: Yann Boursiac, Christophe Pradal, Fabrice Bauget, Mikaël Lucas, Stathis Delivorias, Christophe Godin, Christophe Maurel, Institut des Sciences des Plantes de Montpellier (IPSIM), Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro Montpellier, Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Université de Montpellier (UM), Amélioration génétique et adaptation des plantes méditerranéennes et tropicales (UMR AGAP), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Agro Montpellier, Département Systèmes Biologiques (Cirad-BIOS), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad), Scientific Data Management (ZENITH), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Simulation et Analyse de la morphogenèse in siliCo (MOSAIC), Inria Lyon, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Zenith, and Mosaic
Subjects: Arabidopsis thaliana, Physiology, [SDV]Life Sciences [q-bio], Arabidopsis, Phénotype, Conductance hydraulique des racines, Plant Science, Aquaporins, Plant Roots, Relation plante eau, Système racinaire, Xylem, Genetics, [SDV.BV]Life Sciences [q-bio]/Vegetal Biology, Absorption d'eau, U10 - Informatique, mathématiques et statistiques, Water, Biological Transport, Physiologie végétale, F61 - Physiologie végétale - Nutrition, Modélisation
Abstract: Water uptake by roots is a key adaptation of plants to aerial life. Water uptake depends on root system architecture (RSA) and tissue hydraulic properties that, together, shape the root hydraulic architecture. This work investigates how the interplay between conductivities along radial (e.g. aquaporins) and axial (e.g. xylem vessels) pathways determines the water transport properties of highly branched RSAs as found in adult Arabidopsis (Arabidopsis thaliana) plants. A hydraulic model named HydroRoot was developed, based on multi-scale tree graph representations of RSAs. Root water flow was measured by the pressure chamber technique after successive cuts of a same root system from the tip toward the base. HydroRoot model inversion in corresponding RSAs allowed us to concomitantly determine radial and axial conductivities, providing evidence that the latter is often overestimated by classical evaluation based on the Hagen–Poiseuille law. Organizing principles of Arabidopsis primary and lateral root growth and branching were determined and used to apply the HydroRoot model to an extended set of simulated RSAs. Sensitivity analyses revealed that water transport can be co-limited by radial and axial conductances throughout the whole RSA. The number of roots that can be sectioned (intercepted) at a given distance from the base was defined as an accessible and informative indicator of RSA. The overall set of experimental and theoretical procedures was applied to plants mutated in ESKIMO1 and previously shown to have xylem collapse. This approach will be instrumental to dissect the root water transport phenotype of plants with intricate alterations in root growth or transport functions.
Published: 2022

Catalog

Books, media, physical & digital resources

See catalog results

Searchworks

Select search scope, currently: Articles Catalog books, media & more in Jio Institute collections Articles journal articles & other e-resources

Search

Search Constraints

Refine your results

Search Limiters

Topic

Publication Year Range

Language

Publication Type

Journal

Database

Publisher

615 results on '"Scientific Data Management (ZENITH)"'

Search Results

Catalog

Select search scope, currently: Articles

Catalog

books, media & more in Jio Institute collections

Articles

journal articles & other e-resources