Search

Reading comprehension continues to be a crucial research focus in the NLP community. Recent advances in Machine Reading Comprehension (MRC) have mostly centered on literal comprehension, referring to the surface-level understanding of content. In this work, we focus on the next level - interpretive comprehension, with a particular emphasis on inferring the themes of a narrative text. We introduce the first dataset specifically designed for interpretive comprehension of educational narratives, providing corresponding well-edited theme texts. The dataset spans a variety of genres and cultural origins and includes human-annotated theme keywords with varying levels of granularity. We further formulate NLP tasks under different abstractions of interpretive comprehension toward the main idea of a story. After conducting extensive experiments with state-of-the-art methods, we found the task to be both challenging and significant for NLP research. The dataset and source code have been made publicly available to the research community at https://github.com/RiTUAL-UH/EduStory., Comment: Accepted at LREC-COLING 2024 (long paper)

Machine Learning (ML) is being used in multiple disciplines due to its powerful capability to infer relationships within data. In particular, Software Engineering (SE) is one of those disciplines in which ML has been used for multiple tasks, like software categorization, bugs prediction, and testing. In addition to the multiple ML applications, some studies have been conducted to detect and understand possible pitfalls and issues when using ML. However, to the best of our knowledge, only a few studies have focused on presenting ML best practices or guidelines for the application of ML in different domains. In addition, the practices and literature presented in previous literature (i) are domain-specific (e.g., concrete practices in biomechanics), (ii) describe few practices, or (iii) the practices lack rigorous validation and are presented in gray literature. In this paper, we present a study listing 127 ML best practices systematically mining 242 posts of 14 different Stack Exchange (STE) websites and validated by four independent ML experts. The list of practices is presented in a set of categories related to different stages of the implementation process of an ML-enabled system; for each practice, we include explanations and examples. In all the practices, the provided examples focus on SE tasks. We expect this list of practices could help practitioners to understand better the practices and use ML in a more informed way, in particular newcomers to this new area that sits at the intersection of software engineering and machine learning.

This paper introduces a novel approach to probabilistic deep learning, kernel density matrices, which provide a simpler yet effective mechanism for representing joint probability distributions of both continuous and discrete random variables. In quantum mechanics, a density matrix is the most general way to describe the state of a quantum system. This work extends the concept of density matrices by allowing them to be defined in a reproducing kernel Hilbert space. This abstraction allows the construction of differentiable models for density estimation, inference, and sampling, and enables their integration into end-to-end deep neural models. In doing so, we provide a versatile representation of marginal and joint probability distributions that allows us to develop a differentiable, compositional, and reversible inference procedure that covers a wide range of machine learning tasks, including density estimation, discriminative learning, and generative modeling. The broad applicability of the framework is illustrated by two examples: an image classification model that can be naturally transformed into a conditional generative model, and a model for learning with label proportions that demonstrates the framework's ability to deal with uncertainty in the training samples. The framework is implemented as a library and is available at: https://github.com/fagonzalezo/kdm.

Despite major advances in artificial intelligence (AI) for medicine and healthcare, the deployment and adoption of AI technologies remain limited in real-world clinical practice. In recent years, concerns have been raised about the technical, clinical, ethical and legal risks associated with medical AI. To increase real world adoption, it is essential that medical AI tools are trusted and accepted by patients, clinicians, health organisations and authorities. This work describes the FUTURE-AI guideline as the first international consensus framework for guiding the development and deployment of trustworthy AI tools in healthcare. The FUTURE-AI consortium was founded in 2021 and currently comprises 118 inter-disciplinary experts from 51 countries representing all continents, including AI scientists, clinicians, ethicists, and social scientists. Over a two-year period, the consortium defined guiding principles and best practices for trustworthy AI through an iterative process comprising an in-depth literature review, a modified Delphi survey, and online consensus meetings. The FUTURE-AI framework was established based on 6 guiding principles for trustworthy AI in healthcare, i.e. Fairness, Universality, Traceability, Usability, Robustness and Explainability. Through consensus, a set of 28 best practices were defined, addressing technical, clinical, legal and socio-ethical dimensions. The recommendations cover the entire lifecycle of medical AI, from design, development and validation to regulation, deployment, and monitoring. FUTURE-AI is a risk-informed, assumption-free guideline which provides a structured approach for constructing medical AI tools that will be trusted, deployed and adopted in real-world practice. Researchers are encouraged to take the recommendations into account in proof-of-concept stages to facilitate future translation towards clinical practice of medical AI.

Despite major advances in artificial intelligence (AI) for medicine and healthcare, the deployment and adoption of AI technologies remain limited in real-world clinical practice. In recent years, concerns have been raised about the technical, clinical, ethical and legal risks associated with medical AI. To increase real world adoption, it is essential that medical AI tools are trusted and accepted by patients, clinicians, health organisations and authorities. This work describes the FUTURE-AI guideline as the first international consensus framework for guiding the development and deployment of trustworthy AI tools in healthcare. The FUTURE-AI consortium was founded in 2021 and currently comprises 118 inter-disciplinary experts from 51 countries representing all continents, including AI scientists, clinicians, ethicists, and social scientists. Over a two-year period, the consortium defined guiding principles and best practices for trustworthy AI through an iterative process comprising an in-depth literature review, a modified Delphi survey, and online consensus meetings. The FUTURE-AI framework was established based on 6 guiding principles for trustworthy AI in healthcare, i.e. Fairness, Universality, Traceability, Usability, Robustness and Explainability. Through consensus, a set of 28 best practices were defined, addressing technical, clinical, legal and socio-ethical dimensions. The recommendations cover the entire lifecycle of medical AI, from design, development and validation to regulation, deployment, and monitoring. FUTURE-AI is a risk-informed, assumption-free guideline which provides a structured approach for constructing medical AI tools that will be trusted, deployed and adopted in real-world practice. Researchers are encouraged to take the recommendations into account in proof-of-concept stages to facilitate future translation towards clinical practice of medical AI.

In this work, we introduce a pioneering research challenge: evaluating positive and potentially harmful messages within music products. We initiate by setting a multi-faceted, multi-task benchmark for music content assessment. Subsequently, we introduce an efficient multi-task predictive model fortified with ordinality-enforcement to address this challenge. Our findings reveal that the proposed method not only significantly outperforms robust task-specific alternatives but also possesses the capability to assess multiple aspects simultaneously. Furthermore, through detailed case studies, where we employed Large Language Models (LLMs) as surrogates for content assessment, we provide valuable insights to inform and guide future research on this topic. The code for dataset creation and model implementation is publicly available at https://github.com/RiTUAL-UH/music-message-assessment., Comment: Accepted at LREC-COLING 2024 (long paper)

This work addresses the challenge of training supervised machine or deep learning models on orbiting platforms where we are generally constrained by limited on-board hardware capabilities and restricted uplink bandwidths to upload. We aim at enabling orbiting spacecrafts to (1) continuously train a lightweight model as it acquires imagery; and (2) receive new labels while on orbit to refine or even change the predictive task being trained. For this, we consider chip level regression tasks (i.e. predicting the vegetation percentage of a 20 km$^2$ patch) when we only have coarser label proportions, such as municipality level vegetation statistics (a municipality containing several patches). Such labels proportions have the additional advantage that usually come in tabular data and are widely available in many regions of the world and application areas. This can be framed as a Learning from Label Proportions (LLP) problem setup. LLP applied to Earth Observation (EO) data is still an emerging field and performing comparative studies in applied scenarios remains a challenge due to the lack of standardized datasets. In this work, first, we show how very simple deep learning and probabilistic methods (with {\raise.17ex\hbox{$\scriptstyle\sim$}}5K parameters) generally perform better than standard more complex ones, providing a surprising level of finer grained spatial detail when trained with much coarser label proportions. Second, we publish a set of benchmarking datasets enabling comparative LLP applied to EO, providing both fine grained labels and aggregated data according to existing administrative divisions. Finally, we show how this approach fits an on-orbit training scenario by reducing vastly both the amount of computing and the size of the labels sets. Source code is available at https://github.com/rramosp/llpeo, Comment: 16 pages, 13 figures

This paper presents an anomaly detection model that combines the strong statistical foundation of density-estimation-based anomaly detection methods with the representation-learning ability of deep-learning models. The method combines an autoencoder, for learning a low-dimensional representation of the data, with a density-estimation model based on random Fourier features and density matrices in an end-to-end architecture that can be trained using gradient-based optimization techniques. The method predicts a degree of normality for new samples based on the estimated density. A systematic experimental evaluation was performed on different benchmark datasets. The experimental results show that the method performs on par with or outperforms other state-of-the-art methods., Comment: 10 pages

This paper presents a novel density estimation method for anomaly detection using density matrices (a powerful mathematical formalism from quantum mechanics) and Fourier features. The method can be seen as an efficient approximation of Kernel Density Estimation (KDE). A systematic comparison of the proposed method with eleven state-of-the-art anomaly detection methods on various data sets is presented, showing competitive performance on different benchmark data sets. The method is trained efficiently and it uses optimization to find the parameters of data embedding. The prediction phase complexity of the proposed algorithm is constant relative to the training data size, and it performs well in data sets with different anomaly rates. Its architecture allows vectorization and can be implemented on GPU/TPU hardware., Comment: 10 pages, 1 figure

Diabetic retinopathy (DR) is one of the leading causes of blindness in the working-age population of developed countries, caused by a side effect of diabetes that reduces the blood supply to the retina. Deep neural networks have been widely used in automated systems for DR classification on eye fundus images. However, these models need a large number of annotated images. In the medical domain, annotations from experts are costly, tedious, and time-consuming; as a result, a limited number of annotated images are available. This paper presents a semi-supervised method that leverages unlabeled images and labeled ones to train a model that detects diabetic retinopathy. The proposed method uses unsupervised pretraining via self-supervised learning followed by supervised fine-tuning with a small set of labeled images and knowledge distillation to increase the performance in classification task. This method was evaluated on the EyePACS test and Messidor-2 dataset achieving 0.94 and 0.89 AUC respectively using only 2% of EyePACS train labeled images.

Kernel density estimation (KDE) is one of the most widely used nonparametric density estimation methods. The fact that it is a memory-based method, i.e., it uses the entire training data set for prediction, makes it unsuitable for most current big data applications. Several strategies, such as tree-based or hashing-based estimators, have been proposed to improve the efficiency of the kernel density estimation method. The novel density kernel density estimation method (DMKDE) uses density matrices, a quantum mechanical formalism, and random Fourier features, an explicit kernel approximation, to produce density estimates. This method has its roots in the KDE and can be considered as an approximation method, without its memory-based restriction. In this paper, we systematically evaluate the novel DMKDE algorithm and compare it with other state-of-the-art fast procedures for approximating the kernel density estimation method on different synthetic data sets. Our experimental results show that DMKDE is on par with its competitors for computing density estimates and advantages are shown when performed on high-dimensional data. We have made all the code available as an open source software repository., Comment: 9 pages, 3 figures, 1 table

Density estimation is a fundamental task in statistics and machine learning applications. Kernel density estimation is a powerful tool for non-parametric density estimation in low dimensions; however, its performance is poor in higher dimensions. Moreover, its prediction complexity scale linearly with more training data points. This paper presents a method for neural density estimation that can be seen as a type of kernel density estimation, but without the high prediction computational complexity. The method is based on density matrices, a formalism used in quantum mechanics, and adaptive Fourier features. The method can be trained without optimization, but it could be also integrated with deep learning architectures and trained using gradient descent. Thus, it could be seen as a form of neural density estimation method. The method was evaluated in different synthetic and real datasets, and its performance compared against state-of-the-art neural density estimation methods, obtaining competitive results., Comment: 14 pages, 6 figures, 2 tables

We demonstrate the implementation of a novel machine learning framework for probability density estimation and classification using quantum circuits. The framework maps a training data set or a single data sample to the quantum state of a physical system through quantum feature maps. The quantum state of the arbitrarily large training data set summarises its probability distribution in a finite-dimensional quantum wave function. By projecting the quantum state of a new data sample onto the quantum state of the training data set, one can derive statistics to classify or estimate the density of the new data sample. Remarkably, the implementation of our framework on a real quantum device does not require any optimisation of quantum circuit parameters. Nonetheless, we discuss a variational quantum circuit approach that could leverage quantum advantage for our framework., Comment: Paper condensing experiments shown in QTML 2021

We consider the Feynman-Kitaev formalism applied to a spin chain described by the transverse field Ising model. This formalism consists of building a Hamiltonian whose ground state encodes the time evolution of the spin chain at discrete time steps. To find this ground state, variational wave functions parameterised by artificial neural networks -- also known as neural quantum states (NQSs) -- are used. Our work focuses on assessing, in the context of the Feynman-Kitaev formalism, two properties of NQSs: expressivity (the possibility that variational parameters can be set to values such that the NQS is faithful to the true ground state of the system) and trainability (the process of reaching said values). We find that the considered NQSs are capable of accurately approximating the true ground state of the system, i.e., they are expressive enough ans\"atze. However, extensive hyperparameter tuning experiments show that, empirically, reaching the set of values for the variational parameters that correctly describe the ground state becomes ever more difficult as the number of time steps increase because the true ground state becomes more entangled, and the probability distribution starts to spread across the Hilbert space canonical basis., Comment: 20 pages, 4 figures

QCD jets produced in heavy-ion collisions at LHC or RHIC energies partially evolve inside the produced hot and dense quark gluon plasma, offering unique opportunities to study QCD splitting processes in different backgrounds. Induced (modified) splittings are expected to be the most relevant mechanism driving the modifications of in-medium jets compared to vacuum jets for a wide sets of observables. Although color coherence among different emitters has been identified as an essential mechanism in studies of the QCD antenna radiation, it is usually neglected in the multi-gluon medium-induced cascade. This independent gluon emission approximation can be analytically proved to be valid in the limit of very large media, but corrections or modifications to it have not been computed before in the context of the evolution (or rate) equation describing the gluon cascade. We propose a modified evolution equation that includes corrections due to the interference of subsequent emitters. In order to do so, we first compute a modified splitting kernel following the usual procedure of factorizing it from the subsequent Brownian motion. The calculation is performed in the two-gluon configuration with no overlapping formation times, that is expected to provide the first correction to the completely independent picture

Diabetic retinopathy (DR) is the result of a complication of diabetes affecting the retina. It can cause blindness, if left undiagnosed and untreated. An ophthalmologist performs the diagnosis by screening each patient and analyzing the retinal lesions via ocular imaging. In practice, such analysis is time-consuming and cumbersome to perform. This paper presents a model for automatic DR classification on eye fundus images. The approach identifies the main ocular lesions related to DR and subsequently diagnoses the illness. The proposed method follows the same workflow as the clinicians, providing information that can be interpreted clinically to support the prediction. A subset of the kaggle EyePACS and the Messidor-2 datasets, labeled with ocular lesions, is made publicly available. The kaggle EyePACS subset is used as a training set and the Messidor-2 as a test set for lesions and DR classification models. For DR diagnosis, our model has an area-under-the-curve, sensitivity, and specificity of 0.948, 0.886, and 0.875, respectively, which competes with state-of-the-art approaches., Comment: 7 pages and 1 figure

This paper presents a hybrid classical-quantum program for density estimation and supervised classification. The program is implemented as a quantum circuit in a high-dimensional quantum computer simulator. We show that the proposed quantum protocols allow to estimate probability density functions and to make predictions in a supervised learning manner. This model can be generalized to find expected values of density matrices in high-dimensional quantum computers. Experiments on various data sets are presented. Results show that the proposed method is a viable strategy to implement supervised classification and density estimation in a high-dimensional quantum computer., Comment: 15 pages, 10 figures

Prostate cancer (PCa) is one of the most common and aggressive cancers worldwide. The Gleason score (GS) system is the standard way of classifying prostate cancer and the most reliable method to determine the severity and treatment to follow. The pathologist looks at the arrangement of cancer cells in the prostate and assigns a score on a scale that ranges from 6 to 10. Automatic analysis of prostate whole-slide images (WSIs) is usually addressed as a binary classification problem, which misses the finer distinction between stages given by the GS. This paper presents a probabilistic deep learning ordinal classification method that can estimate the GS from a prostate WSI. Approaching the problem as an ordinal regression task using a differentiable probabilistic model not only improves the interpretability of the results, but also improves the accuracy of the model when compared to conventional deep classification and regression architectures.

We present a map from the travelling salesman problem (TSP), a prototypical NP-complete combinatorial optimisation task, to the ground state associated with a system of many-qudits. Conventionally, the TSP is cast into a quadratic unconstrained binary optimisation (QUBO) problem, that can be solved on an Ising machine. The size of the corresponding physical system's Hilbert space is $2^{N^2}$, where $N$ is the number of cities considered in the TSP. Our proposal provides a many-qudit system with a Hilbert space of dimension $2^{N\log_2N}$, which is considerably smaller than the dimension of the Hilbert space of the system resulting from the usual QUBO map. This reduction can yield a significant speedup in quantum and classical computers. We simulate and validate our proposal using variational Monte Carlo with a neural quantum state, solving the TSP in a linear layout for up to almost 100 cities., Comment: 9 pages, 4 figures

A density matrix describes the statistical state of a quantum system. It is a powerful formalism to represent both the quantum and classical uncertainty of quantum systems and to express different statistical operations such as measurement, system combination and expectations as linear algebra operations. This paper explores how density matrices can be used as a building block for machine learning models exploiting their ability to straightforwardly combine linear algebra and probability. One of the main results of the paper is to show that density matrices coupled with random Fourier features could approximate arbitrary probability distributions over $\mathbb{R}^n$. Based on this finding the paper builds different models for density estimation, classification and regression. These models are differentiable, so it is possible to integrate them with other differentiable components, such as deep learning architectures and to learn their parameters using gradient-based optimization. In addition, the paper presents optimization-less training strategies based on estimation and model averaging. The models are evaluated in benchmark tasks and the results are reported and discussed., Comment: Final version published in Quantum Mach. Intell. 4, 23 (2022)

[EN] Passage retrieval is an important stage of question answering systems. Closed domain passage retrieval, e.g. biomedical passage retrieval presents additional challenges such as specialized terminology, more complex and elaborated queries, scarcity in the amount of available data, among others. However, closed domains also offer some advantages such as the availability of specialized structured information sources, e.g. ontologies and thesauri, that could be used to improve retrieval performance. This paper presents a novel approach for biomedical passage retrieval which is able to combine different information sources using a similarity matrix fusion strategy based on convolutional neural network architecture. The method was evaluated over the standard BioASQ dataset, a dataset specialized on biomedical question answering. The results show that the method is an effective strategy for biomedical passage retrieval able to outperform other state-of-the-art methods in this domain.

[EN] Passage retrieval is an important stage of question answering systems. Closed domain passage retrieval, e.g. biomedical passage retrieval presents additional challenges such as specialized terminology, more complex and elaborated queries, scarcity in the amount of available data, among others. However, closed domains also offer some advantages such as the availability of specialized structured information sources, e.g. ontologies and thesauri, that could be used to improve retrieval performance. This paper presents a novel approach for biomedical passage retrieval which is able to combine different information sources using a similarity matrix fusion strategy based on convolutional neural network architecture. The method was evaluated over the standard BioASQ dataset, a dataset specialized on biomedical question answering. The results show that the method is an effective strategy for biomedical passage retrieval able to outperform other state-of-the-art methods in this domain.

Presentamos los cuentos y dibujos destacados del Festival de Creación Literaria para niños y jóvenes, organizado por la Universidad de los niños de la Universidad EAFIT, el Centro Cultural Biblioteca Luis Echavarría Villegas y el pregrado en Literatura del Departamento de Humanidades., Universidad de los niños de la Universidad EAFIT, el Centro Cultural Biblioteca Luis Echavarría Villegas y el pregrado en Literatura del Departamento de Humanidades.

Diabetic Retinopathy (DR) is one of the microvascular complications of Diabetes Mellitus, which remains as one of the leading causes of blindness worldwide. Computational models based on Convolutional Neural Networks represent the state of the art for the automatic detection of DR using eye fundus images. Most of the current work address this problem as a binary classification task. However, including the grade estimation and quantification of predictions uncertainty can potentially increase the robustness of the model. In this paper, a hybrid Deep Learning-Gaussian process method for DR diagnosis and uncertainty quantification is presented. This method combines the representational power of deep learning, with the ability to generalize from small datasets of Gaussian process models. The results show that uncertainty quantification in the predictions improves the interpretability of the method as a diagnostic support tool. The source code to replicate the experiments is publicly available at https://github.com/stoledoc/DLGP-DR-Diagnosis.

The dissimilarity mixture autoencoder (DMAE) is a neural network model for feature-based clustering that incorporates a flexible dissimilarity function and can be integrated into any kind of deep learning architecture. It internally represents a dissimilarity mixture model (DMM) that extends classical methods like K-Means, Gaussian mixture models, or Bregman clustering to any convex and differentiable dissimilarity function through the reinterpretation of probabilities as neural network representations. DMAE can be integrated with deep learning architectures into end-to-end models, allowing the simultaneous estimation of the clustering and neural network's parameters. Experimental evaluation was performed on image and text clustering benchmark datasets showing that DMAE is competitive in terms of unsupervised classification accuracy and normalized mutual information. The source code with the implementation of DMAE is publicly available at: https://github.com/juselara1/dmae, Comment: 20 pages (2.5 pages of references)

This paper reports a novel method for supervised machine learning based on the mathematical formalism that supports quantum mechanics. The method uses projective quantum measurement as a way of building a prediction function. Specifically, the relationship between input and output variables is represented as the state of a bipartite quantum system. The state is estimated from training samples through an averaging process that produces a density matrix. Prediction of the label for a new sample is made by performing a projective measurement on the bipartite system with an operator, prepared from the new input sample, and applying a partial trace to obtain the state of the subsystem representing the output. The method can be seen as a generalization of Bayesian inference classification and as a type of kernel-based learning method. One remarkable characteristic of the method is that it does not require learning any parameters through optimization. We illustrate the method with different 2-D classification benchmark problems and different quantum information encodings., Comment: Supplementary material integrated into main text. Typos corrected

In this article, we study the relation extraction problem from Natural Language Processing (NLP) implementing a domain adaptation setting without external resources. We trained a Deep Learning (DL) model for Relation Extraction (RE), which extracts semantic relations in the biomedical domain. However, can the model be applied to different domains? The model should be adaptable to automatically extract relationships across different domains using the DL network. Completely training DL models in a short time is impractical because the models should quickly adapt to different datasets in several domains without delay. Therefore, adaptation is crucial for intelligent systems, where changing factors and unanticipated perturbations are common. In this study, we present a detailed analysis of the problem, as well as preliminary experimentation, results, and their evaluation., En este trabajo estudiamos el problema de extracción de relaciones del Procesamiento de Lenguaje Natural (PLN). Realizamos una configuración para la adaptación de dominio sin recursos externos. De esta forma, entrenamos un modelo con aprendizaje profundo (DL) para la extracción de relaciones (RE). El modelo permite extraer relaciones semánticas para el dominio biomédico. Sin embargo, ¿El modelo puede ser aplicado a diferentes dominios? El modelo debería adaptarse automáticamente para la extracción de relaciones entre diferentes dominios usando la red de DL. Entrenar completamente modelos DL en una escala de tiempo corta no es práctico, deseamos que los modelos se adapten rápidamente de diferentes conjuntos de datos con varios dominios y sin demora. Así, la adaptación es crucial para los sistemas inteligentes que operan en el mundo real, donde los factores cambiantes y las perturbaciones imprevistas son habituales. En este artículo, presentamos un análisis detallado del problema, una experimentación preliminar, resultados y la discusión acerca de los resultados.

In this article, we study the relation extraction problem from Natural Language Processing (NLP) implementing a domain adaptation setting without external resources. We trained a Deep Learning (DL) model for Relation Extraction (RE), which extracts semantic relations in the biomedical domain. However, can the model be applied to different domains? The model should be adaptable to automatically extract relationships across different domains using the DL network. Completely training DL models in a short time is impractical because the models should quickly adapt to different datasets in several domains without delay. Therefore, adaptation is crucial for intelligent systems, where changing factors and unanticipated perturbations are common. In this study, we present a detailed analysis of the problem, as well as preliminary experimentation, results, and their evaluation., En este trabajo estudiamos el problema de extracción de relaciones del Procesamiento de Lenguaje Natural (PLN). Realizamos una configuración para la adaptación de dominio sin recursos externos. De esta forma, entrenamos un modelo con aprendizaje profundo (DL) para la extracción de relaciones (RE). El modelo permite extraer relaciones semánticas para el dominio biomédico. Sin embargo, ¿El modelo puede ser aplicado a diferentes dominios? El modelo debería adaptarse automáticamente para la extracción de relaciones entre diferentes dominios usando la red de DL. Entrenar completamente modelos DL en una escala de tiempo corta no es práctico, deseamos que los modelos se adapten rápidamente de diferentes conjuntos de datos con varios dominios y sin demora. Así, la adaptación es crucial para los sistemas inteligentes que operan en el mundo real, donde los factores cambiantes y las perturbaciones imprevistas son habituales. En este artículo, presentamos un análisis detallado del problema, una experimentación preliminar, resultados y la discusión acerca de los resultados.

Recognizing named entities in a document is a key task in many NLP applications. Although current state-of-the-art approaches to this task reach a high performance on clean text (e.g. newswire genres), those algorithms dramatically degrade when they are moved to noisy environments such as social media domains. We present two systems that address the challenges of processing social media data using character-level phonetics and phonology, word embeddings, and Part-of-Speech tags as features. The first model is a multitask end-to-end Bidirectional Long Short-Term Memory (BLSTM)-Conditional Random Field (CRF) network whose output layer contains two CRF classifiers. The second model uses a multitask BLSTM network as feature extractor that transfers the learning to a CRF classifier for the final prediction. Our systems outperform the current F1 scores of the state of the art on the Workshop on Noisy User-generated Text 2017 dataset by 2.45% and 3.69%, establishing a more suitable approach for social media environments., Comment: NAACL 2018

The main goal of this paper is to explore latent topic analysis (LTA), in the context of quantum information retrieval. LTA is a valuable technique for document analysis and representation, which has been extensively used in information retrieval and machine learning. Different LTA techniques have been proposed, some based on geometrical modeling (such as latent semantic analysis, LSA) and others based on a strong statistical foundation. However, these two different approaches are not usually mixed. Quantum information retrieval has the remarkable virtue of combining both geometry and probability in a common principled framework. We built on this quantum framework to propose a new LTA method, which has a clear geometrical motivation but also supports a well-founded probabilistic interpretation. An initial exploratory experimentation was performed on three standard data sets. The results show that the proposed method outperforms LSA on two of the three datasets. These results suggests that the quantum-motivated representation is an alternative for geometrical latent topic modeling worthy of further exploration., Comment: ICTIR2011 International Conference on the Theory of Information Retrieval

This paper presents a novel model for multimodal learning based on gated neural networks. The Gated Multimodal Unit (GMU) model is intended to be used as an internal unit in a neural network architecture whose purpose is to find an intermediate representation based on a combination of data from different modalities. The GMU learns to decide how modalities influence the activation of the unit using multiplicative gates. It was evaluated on a multilabel scenario for genre classification of movies using the plot and the poster. The GMU improved the macro f-score performance of single-modality approaches and outperformed other fusion strategies, including mixture of experts models. Along with this work, the MM-IMDb dataset is released which, to the best of our knowledge, is the largest publicly available multimodal dataset for genre prediction on movies.

La cocaína es una de las drogas de abuso más habituales entre el consumo de la población a nivel mundial. Presenta dos formas diferentes que generan los mismos síntomas en el organismo, la cocaína protonada (C17H22NO4+Cl-) (presentada habitualmente en forma de clorohidrato), y la cocaína base o “crack” (C17H21NO4). La mayoría de los test empleados en la detección de estas drogas se basan en la reacción con un complejo de cobalto (II). En la actualidad, se desconoce con certeza la forma por la que la cocaína se une a este complejo y la geometría del producto final. Por este motivo, en este trabajo se ha realizado un primer estudio sobre la estabilidad de los diferentes confórmeros de cada una de las estructuras, con la finalidad de obtener una primera base para estudios posteriores sobre la formación del complejo entre la cocaína y el compuesto de Co(II). Para realizar este estudio teórico, se han empleado diferentes metodologías computacionales, desde métodos que emplean la mecánica molecular hasta métodos basados en la Teoría del Funcional de la Densidad (DFT). Finalmente se ha realizado un análisis de la densidad electrónica que permita observar las posibles interacciones intramoleculares que ayudan a estabilizar cada uno de los confórmeros obtenidos., Grado en Química

La cocaína es una de las drogas de abuso más habituales entre el consumo de la población a nivel mundial. Presenta dos formas diferentes que generan los mismos síntomas en el organismo, la cocaína protonada (C17H22NO4+Cl-) (presentada habitualmente en forma de clorohidrato), y la cocaína base o “crack” (C17H21NO4). La mayoría de los test empleados en la detección de estas drogas se basan en la reacción con un complejo de cobalto (II). En la actualidad, se desconoce con certeza la forma por la que la cocaína se une a este complejo y la geometría del producto final. Por este motivo, en este trabajo se ha realizado un primer estudio sobre la estabilidad de los diferentes confórmeros de cada una de las estructuras, con la finalidad de obtener una primera base para estudios posteriores sobre la formación del complejo entre la cocaína y el compuesto de Co(II). Para realizar este estudio teórico, se han empleado diferentes metodologías computacionales, desde métodos que emplean la mecánica molecular hasta métodos basados en la Teoría del Funcional de la Densidad (DFT). Finalmente se ha realizado un análisis de la densidad electrónica que permita observar las posibles interacciones intramoleculares que ayudan a estabilizar cada uno de los confórmeros obtenidos., Grado en Química

Background: the prediction of mandibular morphology is important in facial reconstruction for forensic purposes as in orthodontics and maxillofacial surgery. This process has been performed through parametric and linear methods based on Caucasian populations; also, these analyzes are performed on lateral cephalograms, but a prediction from a posteroanterior view is not taken into account. Purpose: to predict through Artificial Neural Networks the mandibular morphology using craniomaxillary measures in posteroanterior radiographs. Methods: 229 standardized posteroanterior radiographs from Colombian young adults of both sexes were collected. Coordinates of craniofacial skeletal landmarks were used to create mandibular and craniomaxillary measures. 17 predictor craniomaxillary input variables were selected, measuring widths, heights and angles. Similarly, 13 mandibular measures were selected to be predicted, considering both the right and left sides. Artificial neural networks were used for the prediction process and it was evaluated by a correlation coefficient using a ridge regression between real value and the predicted value. Results: the results found in the model were significant especially for 5 variables of morphological importance in the forensic field: the right mandibular ramus (Cdd-God), the bigonial width (Goi-God), the bicondylar width (Cdi-Cdd) and the distances between the condyles to the menton (Cdd-Me and Cdi-Me). Conclusions: an important prediction capacity in 5 measures of forensic importance in patients with skeletal Class I, Class II and Class III was found in both sexes., Antecedentes: la predicción de la morfología mandibular es importante tanto en la reconstrucción facial con fines forenses, como en ortodoncia y cirugía maxilofacial. Dicho proceso se ha realizado a través de métodos paramétricos y lineales basándose en poblaciones caucásicas; asimismo, dichos análisis se realizan en radiografías de perfil más no se tiene en cuenta una predicción mandibular desde una vista posteroanterior. Propósito: predecir a través de redes neuronales artificiales la morfología mandibular, empleando medidas craneomaxilares en radiografías posteroanteriores. Métodos: se recolectaron 229 radiografías posteroanteriores estandarizadas de adultos jóvenes colombianos de ambos sexos. Se usaron coordenadas de puntos de referencia óseos craneofaciales para formar medidas mandibulares y craneomaxilares. Se seleccionaron 17 variables predictoras craneomaxilares de entrada, midiendo anchuras, alturas y ángulos, De la misma manera se seleccionaron 13 medidas mandibulares a predecir, considerando tanto el lado derecho como el izquierdo. Se usaron redes neuronales artificiales para realizar el proceso de predicción y se evaluó a través de un coeficiente de correlación, por medio de una regresión de arista (ridge regression) entre el valor real y el valor predicho. Resultados: los resultados encontrados dentro del modelo fueron significativos en especial para 5 variables de importancia morfológica dentro del campo forense: la rama mandibular derecha (Cdd-God), el ancho bigoníaco (Goi-God), el ancho bicondilar (Cdi-Cdd) y las distancias entre los cóndilos al mentón (Cdd-Me y Cdi-Me). Conclusión: se encontró una capacidad de predicción importante en 5 medidas de importancia forense en pacientes Clase I, Clase II y Clase III esquelética en ambos sexos.

La cocaína es una de las drogas de abuso más habituales entre el consumo de la población a nivel mundial. Presenta dos formas diferentes que generan los mismos síntomas en el organismo, la cocaína protonada (C17H22NO4+Cl-) (presentada habitualmente en forma de clorohidrato), y la cocaína base o “crack” (C17H21NO4). La mayoría de los test empleados en la detección de estas drogas se basan en la reacción con un complejo de cobalto (II). En la actualidad, se desconoce con certeza la forma por la que la cocaína se une a este complejo y la geometría del producto final. Por este motivo, en este trabajo se ha realizado un primer estudio sobre la estabilidad de los diferentes confórmeros de cada una de las estructuras, con la finalidad de obtener una primera base para estudios posteriores sobre la formación del complejo entre la cocaína y el compuesto de Co(II). Para realizar este estudio teórico, se han empleado diferentes metodologías computacionales, desde métodos que emplean la mecánica molecular hasta métodos basados en la Teoría del Funcional de la Densidad (DFT). Finalmente se ha realizado un análisis de la densidad electrónica que permita observar las posibles interacciones intramoleculares que ayudan a estabilizar cada uno de los confórmeros obtenidos., Grado en Química

Background: Predicting mandibular morphology is important in facial reconstruction for forensic purposes as in orthodontics and maxillofacial surgery. This process has been performed through parametric and linear methods based on Caucasian populations. Also, these analyzes are performed on lateral cephalograms, but a prediction from a posteroanterior view is not taken into account. Purpose: To predict through artificial neural networks the mandibular morphology using craniomaxillary measures in posteroanterior radiographs. Methods: 229 standardized posteroanterior radiographs from Colombian young adults of both sexes were collected. Coordinates of craniofacial skeletal landmarks were used to create mandibular and craniomaxillary measures. 17 predictor craniomaxillary input variables were selected, measuring widths, heights, and angles. Similarly, 13 mandibular measures were selected to be predicted, considering both the right and left sides. Artificial neural networks were used for the prediction process and it was evaluated by a correlation coefficient using a ridge regression between real value and the predicted value. Results: The results found in the model were significant especially for 5 variables of morphological importance in the forensic field: right mandibular ramus (Cdd-God), bigonial width (Goi-God), bicondylar width (Cdi-Cdd), and distance between the condyles to the menton (Cdd-Me and Cdi-Me). Conclusions: An important prediction capacity in 5 measures of forensic importance in patients with skeletal Class I, Class II and Class III was found in both sexes., Antecedentes: La predicción de la morfología mandibular es importante tanto en la reconstrucción facial con fines forenses, como en ortodoncia y cirugía maxilofacial. Dicho proceso se ha realizado a través de métodos paramétricos y lineales basándose en poblaciones caucásicas. Asimismo, dichos análisis se realizan en radiografías de perfil más no se tiene en cuenta una predicción mandibular desde una vista posteroanterior. Propósito: Predecir, a través de redes neuronales artificiales, la morfología mandibular, empleando medidas craneomaxilares en radiografías posteroanteriores. Métodos: Se recolectaron 229 radiografías postero-anteriores estandarizadas de adultos jóvenes colombianos de ambos sexos. Se usaron coordenadas de puntos de referencia óseos craneofaciales para formar medidas mandibulares y craneomaxilares. Se seleccionaron 17 variables predictoras craneomaxilares de entrada, midiendo anchuras, alturas y ángulos. De la misma manera se seleccionaron 13 medidas mandibulares a predecir, considerando tanto el lado derecho como el izquierdo. Se usaron redes neuronales artificiales para realizar el proceso de predicción y se evaluó a través de un coeficiente de correlación, por medio de una regresión de arista (ridge regression) entre el valor real y el valor predicho. Resultados: Los resultados encontrados dentro del modelo fueron significativos en especial para 5 variables de importancia morfológica dentro del campo forense: la rama mandibular derecha (Cdd-God), el ancho bigoníaco (Goi-God), el ancho bicondilar (Cdi-Cdd) y las distancias entre los cóndilos al mentón (Cdd-Me y Cdi-Me). Conclusión: se encontró una capacidad de predicción importante en 5 medidas de importancia forense en pacientes clase I, clase II y clase III esquelética en ambos sexos.

La cocaína es una de las drogas de abuso más habituales entre el consumo de la población a nivel mundial. Presenta dos formas diferentes que generan los mismos síntomas en el organismo, la cocaína protonada (C17H22NO4+Cl-) (presentada habitualmente en forma de clorohidrato), y la cocaína base o “crack” (C17H21NO4). La mayoría de los test empleados en la detección de estas drogas se basan en la reacción con un complejo de cobalto (II). En la actualidad, se desconoce con certeza la forma por la que la cocaína se une a este complejo y la geometría del producto final. Por este motivo, en este trabajo se ha realizado un primer estudio sobre la estabilidad de los diferentes confórmeros de cada una de las estructuras, con la finalidad de obtener una primera base para estudios posteriores sobre la formación del complejo entre la cocaína y el compuesto de Co(II). Para realizar este estudio teórico, se han empleado diferentes metodologías computacionales, desde métodos que emplean la mecánica molecular hasta métodos basados en la Teoría del Funcional de la Densidad (DFT). Finalmente se ha realizado un análisis de la densidad electrónica que permita observar las posibles interacciones intramoleculares que ayudan a estabilizar cada uno de los confórmeros obtenidos., Grado en Química

Personality traits influence most, if not all, of the human activities, from those as natural as the way people walk, talk, dress and write to those most complex as the way they interact with others. Most importantly, personality influences the way people make decisions including, in the case of developers, the criteria they consider when selecting a software project they want to participate. Most of the works that study the influence of social, technical and human factors in software development projects have been focused on the impact of communications in software quality. For instance, on identifying predictors to detect files that may contain bugs before releasing an enhanced version of a software product. Only a few of these works focus on the analysis of personality traits of developers with commit permissions (committers) in Free/Libre and Open-Source Software projects and their relationship with the software artifacts they interact with. This paper presents an approach, based on the automatic recognition of personality traits from e-mails sent by committers in FLOSS projects, to uncover relationships between the social and technical aspects that occur during the software development process. Our experimental results suggest the existence of some relationships among personality traits projected by the committers through their e-mails and the social (communication) and technical activities they undertake. This work is a preliminary study aimed at supporting the setting up of efficient work teams in software development projects based on an appropriate mix of stakeholders taking into account their personality traits., Comment: 7 pages, ICSE 2016 - Workshop on Cooperative and Human Aspects of Software Engineering (CHASE)

Anatomical variability of patient's brains limits the statistical analyses about presence or absence of a pathology. In this paper, we present an approach for classification of brain Magnetic Resonance (MR) images from healthy and diseased subjects. The approach builds up a saliency map, which extract regions of relative change in three different dimensions: intensity, orientation and edges. The obtained regions of interest are used as suitable patterns for subject classification using support vector machines. The strategy’s performance was assessed on a set of 198 MR images extracted from the OASIS database and divided into four groups, reporting an average accuracy rate of 74.54% and an average Equal Error Rate of 0.725.

Anatomical variability of patient's brains limits the statistical analyses about presence or absence of a pathology. In this paper, we present an approach for classification of brain Magnetic Resonance (MR) images from healthy and diseased subjects. The approach builds up a saliency map, which extract regions of relative change in three different dimensions: intensity, orientation and edges. The obtained regions of interest are used as suitable patterns for subject classification using support vector machines. The strategy's performance was assessed on a set of 198 MR images extracted from the OASIS database and divided into four groups, reporting an average accuracy rate of 74.54% and an average Equal Error Rate of 0.725, La variabilidad anatómica presente en los cerebros de pacientes limita la realización de análisis estadísticos acerca de la presencia o ausencia de una patología. En este artículo, presentamos una aproximación para la clasificación de imágenes de Resonancia Magnética (MR) cerebral de sujetos sanos y afectados por una patología. El enfoque se basa en un mapa de saliencia, el cual extrae regiones de cambio relativo en tres diferentes dimensiones: intensidad, orientación y bordes. Las regiones de interés obtenidas se utilizan como patrones para la clasificación de sujetos utilizando máquinas de vectores de soporte. El desempeño de la estrategia propuesta fue evaluado en un conjunto de 198 imágenes de MR extraídas de la base de datos OASIS y divididas en cuatro grupos, reportando una tasa de precisión promedio de 74.54% y una tasa de error igual (Equal Error Rate) promedio de 0.725

Anatomical variability of patient's brains limits the statistical analyses about presence or absence of a pathology. In this paper, we present an approach for classification of brain Magnetic Resonance (MR) images from healthy and diseased subjects. The approach builds up a saliency map, which extract regions of relative change in three different dimensions: intensity, orientation and edges. The obtained regions of interest are used as suitable patterns for subject classification using support vector machines. The strategy’s performance was assessed on a set of 198 MR images extracted from the OASIS database and divided into four groups, reporting an average accuracy rate of 74.54% and an average Equal Error Rate of 0.725.

Anatomical variability of patient's brains limits the statistical analyses about presence or absence of a pathology. In this paper, we present an approach for classification of brain Magnetic Resonance (MR) images from healthy and diseased subjects. The approach builds up a saliency map, which extract regions of relative change in three different dimensions: intensity, orientation and edges. The obtained regions of interest are used as suitable patterns for subject classification using support vector machines. The strategy's performance was assessed on a set of 198 MR images extracted from the OASIS database and divided into four groups, reporting an average accuracy rate of 74.54% and an average Equal Error Rate of 0.725, La variabilidad anatómica presente en los cerebros de pacientes limita la realización de análisis estadísticos acerca de la presencia o ausencia de una patología. En este artículo, presentamos una aproximación para la clasificación de imágenes de Resonancia Magnética (MR) cerebral de sujetos sanos y afectados por una patología. El enfoque se basa en un mapa de saliencia, el cual extrae regiones de cambio relativo en tres diferentes dimensiones: intensidad, orientación y bordes. Las regiones de interés obtenidas se utilizan como patrones para la clasificación de sujetos utilizando máquinas de vectores de soporte. El desempeño de la estrategia propuesta fue evaluado en un conjunto de 198 imágenes de MR extraídas de la base de datos OASIS y divididas en cuatro grupos, reportando una tasa de precisión promedio de 74.54% y una tasa de error igual (Equal Error Rate) promedio de 0.725

Anatomical variability of patient's brains limits the statistical analyses about presence or absence of a pathology. In this paper, we present an approach for classification of brain Magnetic Resonance (MR) images from healthy and diseased subjects. The approach builds up a saliency map, which extract regions of relative change in three different dimensions: intensity, orientation and edges. The obtained regions of interest are used as suitable patterns for subject classification using support vector machines. The strategy's performance was assessed on a set of 198 MR images extracted from the OASIS database and divided into four groups, reporting an average accuracy rate of 74.54% and an average Equal Error Rate of 0.725, La variabilidad anatómica presente en los cerebros de pacientes limita la realización de análisis estadísticos acerca de la presencia o ausencia de una patología. En este artículo, presentamos una aproximación para la clasificación de imágenes de Resonancia Magnética (MR) cerebral de sujetos sanos y afectados por una patología. El enfoque se basa en un mapa de saliencia, el cual extrae regiones de cambio relativo en tres diferentes dimensiones: intensidad, orientación y bordes. Las regiones de interés obtenidas se utilizan como patrones para la clasificación de sujetos utilizando máquinas de vectores de soporte. El desempeño de la estrategia propuesta fue evaluado en un conjunto de 198 imágenes de MR extraídas de la base de datos OASIS y divididas en cuatro grupos, reportando una tasa de precisión promedio de 74.54% y una tasa de error igual (Equal Error Rate) promedio de 0.725

Corpus Linguistics is an empirical methodology which, based on great collections of text - corpus or corpora, attempts to describe the regularities of languages by means of the implementation of computer programs, and in this way to simulate their real use. This work applies Corpus Linguistics to a series of electronic medical histories written in Spanish, which have never been linguistically analyzed before. We do not know the precise form in which these histories were written by doctors or the types of words used when describing an event in a medical subdiscipline. The set of data is formed by 19 medical subdisciplines, which contain their own histories. Each history was tagged in 3 different ways: lemmatization, tokenization, and grammatical part-of-speech, using TreeTagger. Afterwards, the frequencies of tags were described using AntConc. The results found for each subdiscipline show the words that appear with greater frequency. The closed class words were the most commonly used. Some parts of the medical histories were tagged erroneously. On the other hand, examples were found that allowed us to recognize the variability of use of expressions and abbreviations in the medical staff. Also, medical writing at Universidad Nacional de Colombia corroborated Zipf's Law., La Lingüística de Corpus es una metodología empírica ya que, a partir de grandes colecciones de textos -corpus o corpora- intenta describir las regularidades de las lenguas por medio de la implementación de programas computacionales, y así, simular los usos reales de ellas. Este trabajo aplica la Lingüística de Corpus a un conjunto de historias médicas electrónicas escritas en español nunca analizado lingüísticamente. De estas historias se desconoce la forma en que están escritas por parte de los médicos y las clases de palabras que utilizan cuando describen un suceso en una subdisciplina médica. El conjunto de datos está formado por 19 subdisciplinas médicas, las cuales contienen sus propias historias. Cada historia fue anotada en tres formas diferentes, lematización, tokenización y categoría gramatical (part-of-speech) por medio de TreeTagger. Posteriormente, las frecuencias de las anotaciones se describieron mediante AntConc. Los resultados encontrados para cada subdisciplina muestran las palabras con mayor frecuencia. Las palabras de clase cerrada son las más comunes y utilizadas. Algunas partes de las historias médicas fueron anotadas erróneamente. Por otra parte, se muestran ejemplos que dan a conocer la variabilidad de uso entre expresiones y abreviaturas por parte del personal médico. Además, la escritura médica de la Universidad Nacional de Colombia corrobora la Ley de Zipf.

Corpus Linguistics is an empirical methodology which, based on great collections of text - corpus or corpora, attempts to describe the regularities of languages by means of the implementation of computer programs, and in this way to simulate their real use. This work applies Corpus Linguistics to a series of electronic medical histories written in Spanish, which have never been linguistically analyzed before. We do not know the precise form in which these histories were written by doctors or the types of words used when describing an event in a medical subdiscipline. The set of data is formed by 19 medical subdisciplines, which contain their own histories. Each history was tagged in 3 different ways: lemmatization, tokenization, and grammatical part-of-speech, using TreeTagger. Afterwards, the frequencies of tags were described using AntConc. The results found for each subdiscipline show the words that appear with greater frequency. The closed class words were the most commonly used. Some parts of the medical histories were tagged erroneously. On the other hand, examples were found that allowed us to recognize the variability of use of expressions and abbreviations in the medical staff. Also, medical writing at Universidad Nacional de Colombia corroborated Zipf's Law., La Lingüística de Corpus es una metodología empírica ya que, a partir de grandes colecciones de textos -corpus o corpora- intenta describir las regularidades de las lenguas por medio de la implementación de programas computacionales, y así, simular los usos reales de ellas. Este trabajo aplica la Lingüística de Corpus a un conjunto de historias médicas electrónicas escritas en español nunca analizado lingüísticamente. De estas historias se desconoce la forma en que están escritas por parte de los médicos y las clases de palabras que utilizan cuando describen un suceso en una subdisciplina médica. El conjunto de datos está formado por 19 subdisciplinas médicas, las cuales contienen sus propias historias. Cada historia fue anotada en tres formas diferentes, lematización, tokenización y categoría gramatical (part-of-speech) por medio de TreeTagger. Posteriormente, las frecuencias de las anotaciones se describieron mediante AntConc. Los resultados encontrados para cada subdisciplina muestran las palabras con mayor frecuencia. Las palabras de clase cerrada son las más comunes y utilizadas. Algunas partes de las historias médicas fueron anotadas erróneamente. Por otra parte, se muestran ejemplos que dan a conocer la variabilidad de uso entre expresiones y abreviaturas por parte del personal médico. Además, la escritura médica de la Universidad Nacional de Colombia corrobora la Ley de Zipf.

The paper proposes an indexing method for fast content-based retrieval in an image repository. The method is based on a metric-indexing strategy that avoids calculating the distance from the query to all the images in the repository. The proponed indexing method is tested on a pathology image Collection comparing its performance against sequential scanning indexing. The results show that the metric indexing method improves the access time by a factor of 10, without a signi?cant sacr i?ce on precision.

This paper presents an adaptation to the strategy of kernel methods of a wellkn own artificial immune network. This adaptation brings to artificial immune systems, for the first time, some of the advantages of kernel methods, such as the ability to deal with non-vector data and the mapping to high-dimensional spaces through the kernel trick. Preliminar y experiments were carried out in order to get some insights of the behavior of the proposed model.

This paper presents an adaptation to the strategy of kernel methods of a wellkn own artificial immune network. This adaptation brings to artificial immune systems, for the first time, some of the advantages of kernel methods, such as the ability to deal with non-vector data and the mapping to high-dimensional spaces through the kernel trick. Preliminar y experiments were carried out in order to get some insights of the behavior of the proposed model.