Author: "Machine Learning and Optimisation (TAO)" - Searchworks@Jio Institute Digital Library Search Results

Your search keyword '"Machine Learning and Optimisation (TAO)"' showing total 597 results

Start Over Author "Machine Learning and Optimisation (TAO)"

597 results on '"Machine Learning and Optimisation (TAO)"'

1. Benchmarking MATLAB's gamultiobj (NSGA-II) on the Bi-objective BBOB-2016 Test Suite

Author: Tea Tušar, Tobias Wagner, Dejan Tušar, Anne Auger, Nikolaus Hansen, Dimo Brockhoff, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Parallel Cooperative Multi-criteria Optimization (DOLPHIN), 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), Technische Universität Dortmund [Dortmund] (TU), ANR-12-MONU-0009,NumBBO,Analyse, Amelioration, Evaluation d'algorithmes numériques pour l'optimisation boîte-noire(2012), Machine Learning and Optimisation ( TAO ), 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 ) -Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique ( Inria ), Parallel Cooperative Multi-criteria Optimization ( DOLPHIN ), 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 (CRIStAL) - UMR 9189 ( CRIStAL ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Ecole Centrale de Lille-Institut Mines-Télécom [Paris]-Université de Lille-Centre National de la Recherche Scientifique ( CNRS ) -Ecole Centrale de Lille-Institut Mines-Télécom [Paris]-Université de Lille-Centre National de la Recherche Scientifique ( CNRS ), Technische Universität Dortmund [Dortmund] ( TU ), ANR : ANR-2012-MONU-0009,NumBBO, 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), and Université de Lille-Centrale Lille-Centre National de la Recherche Scientifique (CNRS)-Université de Lille-Centrale Lille-Centre National de la Recherche Scientifique (CNRS)
Subjects: [ MATH.MATH-OC ] Mathematics [math]/Optimization and Control [math.OC], Computer science, 0102 computer and information sciences, 02 engineering and technology, Benchmarking, [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], computer.software_genre, 01 natural sciences, Toolbox, 010201 computation theory & mathematics, Black-box optimization, 0202 electrical engineering, electronic engineering, information engineering, Test suite, Bi objective, Benchmark (computing), 020201 artificial intelligence & image processing, Data mining, [MATH.MATH-OC]Mathematics [math]/Optimization and Control [math.OC], [ INFO.INFO-NE ] Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], Bi-objective optimization, MATLAB, computer, computer.programming_language
Abstract: International audience; In this paper, we benchmark a variant of the well-known NSGA-II algorithm of Deb et al. on the biobjective bbob-biobj test suite of the Comparing Continuous Optimizers platform COCO. To this end, we employ the implementation of MATLAB's gamultiobj toolbox with its default settings and a population size of 100.
Published: 2016
Full Text: View/download PDF

2. The Impact of Search Volume on the Performance of RANDOMSEARCH on the Bi-objective BBOB-2016 Test Suite

Author: Dejan Tušar, Nikolaus Hansen, Dimo Brockhoff, Tobias Wagner, Anne Auger, Tea Tušar, Machine Learning and Optimisation (TAO), 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), Parallel Cooperative Multi-criteria Optimization (DOLPHIN), 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), Technische Universität Dortmund [Dortmund] (TU), ANR-12-MONU-0009,NumBBO,Analyse, Amelioration, Evaluation d'algorithmes numériques pour l'optimisation boîte-noire(2012), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, 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 and Optimisation ( TAO ), 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 ) -Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique ( Inria ), Parallel Cooperative Multi-criteria Optimization ( DOLPHIN ), 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 (CRIStAL) - UMR 9189 ( CRIStAL ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Ecole Centrale de Lille-Institut Mines-Télécom [Paris]-Université de Lille-Centre National de la Recherche Scientifique ( CNRS ) -Ecole Centrale de Lille-Institut Mines-Télécom [Paris]-Université de Lille-Centre National de la Recherche Scientifique ( CNRS ), Technische Universität Dortmund [Dortmund] ( TU ), and ANR : ANR-2012-MONU-0009,NumBBO
Subjects: [ MATH.MATH-OC ] Mathematics [math]/Optimization and Control [math.OC], 021103 operations research, 0211 other engineering and technologies, Pareto principle, Sampling (statistics), 02 engineering and technology, [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], Set (abstract data type), Random search, Benchmarking, Dimension (vector space), Search algorithm, Black-box optimization, 0202 electrical engineering, electronic engineering, information engineering, Test suite, 020201 artificial intelligence & image processing, Hypercube, [MATH.MATH-OC]Mathematics [math]/Optimization and Control [math.OC], [ INFO.INFO-NE ] Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], Bi-objective optimization, Algorithm, Mathematics
Abstract: International audience; Pure random search is undeniably the simplest stochastic search algorithm for numerical optimization. Essentially the only thing to be determined to implement the algorithm is its sampling space, the influence of which on the performance on the bi-objective bbob-biobj test suite of the COCO platform is investigated here. It turns out that the suggested region of interest of [−100, 100] n (with n being the problem dimension) has a too vast volume for the algorithm to approximate the Pareto set effectively. Better performance can be achieved if solutions are sampled uniformly within [−5, 5] n or [−4, 4] n. The latter sampling box corresponds to the smallest hypercube encapsulating all single-objective optima of the 55 constructed bi-objective problems of the bbob-biobj test suite. However, not all best known Pareto set approximations are entirely contained within [−5, 5] n .
Published: 2016
Full Text: View/download PDF

3. Benchmarking the Pure Random Search on the Bi-objective BBOB-2016 Testbed

Author: Tobias Wagner, Dimo Brockhoff, Dejan Tušar, Tea Tušar, Anne Auger, Nikolaus Hansen, Machine Learning and Optimisation (TAO), 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), Parallel Cooperative Multi-criteria Optimization (DOLPHIN), 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), Technische Universität Dortmund [Dortmund] (TU), ANR-12-MONU-0009,NumBBO,Analyse, Amelioration, Evaluation d'algorithmes numériques pour l'optimisation boîte-noire(2012), Machine Learning and Optimisation ( TAO ), 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 ) -Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique ( Inria ), Parallel Cooperative Multi-criteria Optimization ( DOLPHIN ), 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 (CRIStAL) - UMR 9189 ( CRIStAL ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Ecole Centrale de Lille-Institut Mines-Télécom [Paris]-Université de Lille-Centre National de la Recherche Scientifique ( CNRS ) -Ecole Centrale de Lille-Institut Mines-Télécom [Paris]-Université de Lille-Centre National de la Recherche Scientifique ( CNRS ), Technische Universität Dortmund [Dortmund] ( TU ), ANR : ANR-2012-MONU-0009,NumBBO, 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, and Université de Lille-Centrale Lille-Centre National de la Recherche Scientifique (CNRS)-Université de Lille-Centrale Lille-Centre National de la Recherche Scientifique (CNRS)
Subjects: [ MATH.MATH-OC ] Mathematics [math]/Optimization and Control [math.OC], 021103 operations research, Computer science, Testbed, 0211 other engineering and technologies, Sampling (statistics), 02 engineering and technology, Benchmarking, [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], computer.software_genre, Computational science, Random search, Dimension (vector space), Black-box optimization, 0202 electrical engineering, electronic engineering, information engineering, Benchmark (computing), Test suite, 020201 artificial intelligence & image processing, Data mining, [MATH.MATH-OC]Mathematics [math]/Optimization and Control [math.OC], [ INFO.INFO-NE ] Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], Bi-objective optimization, computer
Abstract: International audience; The Comparing Continuous Optimizers platform COCO has become a standard for benchmarking numerical (single-objective) optimization algorithms effortlessly. In 2016, COCO has been extended towards multi-objective optimization by providing a first bi-objective test suite. To provide a baseline, we benchmark a pure random search on this bi-objective bbob-biobj test suite of the COCO platform. For each combination of function, dimension n, and instance of the test suite, $10^6 · n$ candidate solutions are sampled uniformly within the sampling box $[−5, 5]^n$ .
Published: 2016
Full Text: View/download PDF

4. The Impact of Variation Operators on the Performance of SMS-EMOA on the Bi-objective BBOB-2016 Test Suite

Author: Tobias Wagner, Dejan Tušar, Tea Tušar, Dimo Brockhoff, Nikolaus Hansen, Anne Auger, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Parallel Cooperative Multi-criteria Optimization (DOLPHIN), 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), Université de Lille-Centrale Lille-Centre National de la Recherche Scientifique (CNRS)-Université de Lille-Centrale Lille-Centre National de la Recherche Scientifique (CNRS), Technische Universität Dortmund [Dortmund] (TU), ANR-12-MONU-0009,NumBBO,Analyse, Amelioration, Evaluation d'algorithmes numériques pour l'optimisation boîte-noire(2012), 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), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), Machine Learning and Optimisation ( TAO ), 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 ) -Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique ( Inria ), Parallel Cooperative Multi-criteria Optimization ( DOLPHIN ), 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 (CRIStAL) - UMR 9189 ( CRIStAL ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Ecole Centrale de Lille-Institut Mines-Télécom [Paris]-Université de Lille-Centre National de la Recherche Scientifique ( CNRS ) -Ecole Centrale de Lille-Institut Mines-Télécom [Paris]-Université de Lille-Centre National de la Recherche Scientifique ( CNRS ), Technische Universität Dortmund [Dortmund] ( TU ), and ANR : ANR-2012-MONU-0009,NumBBO
Subjects: [ MATH.MATH-OC ] Mathematics [math]/Optimization and Control [math.OC], Mathematical optimization, 021103 operations research, Crossover, 0211 other engineering and technologies, Binary number, 02 engineering and technology, Variation (game tree), [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], Benchmarking, Operator (computer programming), Differential evolution, Black-box optimization, 0202 electrical engineering, electronic engineering, information engineering, Benchmark (computing), Test suite, 020201 artificial intelligence & image processing, [MATH.MATH-OC]Mathematics [math]/Optimization and Control [math.OC], [ INFO.INFO-NE ] Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], Bi-objective optimization, Algorithm, Selection (genetic algorithm), Mathematics
Abstract: International audience; The S-metric-Selection Evolutionary Multi-objective Optimization Algorithm (SMS-EMOA) is one of the best-known indicator-based multi-objective optimization algorithms. It employs the S-metric or hypervolume indicator in its (steady-state) selection by deleting in each iteration the solution that has the smallest contribution to the hypervolume indicator. In the SMS-EMOA, the conceptual idea is this hypervolume-based selection. Hence the algorithm can, for example, be combined with several variation operators. Here, we benchmark two versions of SMS-EMOA which employ differential evolution (DE) and simulated binary crossover (SBX) with polynomial mutation (PM) respectively on the newly introduced bi-objective bbob-biobj test suite of the Comparing Continuous Optimizers (COCO) platform. The results un-surprisingly reveal that the choice of the variation operator is crucial for performance with a clear advantage of the DE variant on almost all functions.
Published: 2016
Full Text: View/download PDF

5. Benchmarking RM-MEDA on the Bi-objective BBOB-2016 Test Suite

Author: Nikolaus Hansen, Tea Tušar, Anne Auger, Tobias Wagner, Dimo Brockhoff, Dejan Tušar, Machine Learning and Optimisation (TAO), 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), Parallel Cooperative Multi-criteria Optimization (DOLPHIN), 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), Technische Universität Dortmund [Dortmund] (TU), ANR-12-MONU-0009,NumBBO,Analyse, Amelioration, Evaluation d'algorithmes numériques pour l'optimisation boîte-noire(2012), Machine Learning and Optimisation ( TAO ), 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 ) -Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique ( Inria ), Parallel Cooperative Multi-criteria Optimization ( DOLPHIN ), 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 (CRIStAL) - UMR 9189 ( CRIStAL ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Ecole Centrale de Lille-Institut Mines-Télécom [Paris]-Université de Lille-Centre National de la Recherche Scientifique ( CNRS ) -Ecole Centrale de Lille-Institut Mines-Télécom [Paris]-Université de Lille-Centre National de la Recherche Scientifique ( CNRS ), Technische Universität Dortmund [Dortmund] ( TU ), ANR : ANR-2012-MONU-0009,NumBBO, 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, and Université de Lille-Centrale Lille-Centre National de la Recherche Scientifique (CNRS)-Université de Lille-Centrale Lille-Centre National de la Recherche Scientifique (CNRS)
Subjects: [ MATH.MATH-OC ] Mathematics [math]/Optimization and Control [math.OC], Mathematical optimization, Computer science, 0102 computer and information sciences, 02 engineering and technology, Function (mathematics), Benchmarking, [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], 01 natural sciences, chemistry.chemical_compound, chemistry, Estimation of distribution algorithm, Dimension (vector space), 010201 computation theory & mathematics, Black-box optimization, 0202 electrical engineering, electronic engineering, information engineering, Bi objective, Benchmark (computing), Test suite, 020201 artificial intelligence & image processing, [MATH.MATH-OC]Mathematics [math]/Optimization and Control [math.OC], [ INFO.INFO-NE ] Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], Bi-objective optimization, MEDA
Abstract: International audience; In this paper, we benchmark the Regularity Model-Based Multiobjective Estimation of Distribution Algorithm (RM-MEDA) of Zhang et al. on the bi-objective bbob-biobj test suite of the Comparing Continuous Optimizers (COCO) platform. It turns out that, starting from about 200 times dimension many function evaluations, RM-MEDA shows a linear increase in the solved hypervolume-based target values with time until a stagnation of the performance occurs rather quickly on all problems. The final percentage of solved hy-pervolume targets seems to decrease with the problem dimension .
Published: 2016
Full Text: View/download PDF

6. COCO: Performance Assessment

Author: Hansen, Nikolaus, Auger, Anne, Brockhoff, Dimo, Tušar, Dejan, Tušar, Tea, Machine Learning and Optimisation (TAO), 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), Parallel Cooperative Multi-criteria Optimization (DOLPHIN), Laboratoire d'Informatique Fondamentale de Lille (LIFL), Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS)-Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS)-Inria Lille - Nord Europe, Institut National de Recherche en Informatique et en Automatique (Inria), ANR-12-MONU-0009,NumBBO,Analyse, Amelioration, Evaluation d'algorithmes numériques pour l'optimisation boîte-noire(2012), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Inria Lille - Nord Europe, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique Fondamentale de Lille (LIFL), Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS)-Université de Lille, Sciences et Technologies-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS), Machine Learning and Optimisation ( TAO ), 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 ) -Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique ( Inria ), Parallel Cooperative Multi-criteria Optimization ( DOLPHIN ), Laboratoire d'Informatique Fondamentale de Lille ( LIFL ), Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique ( Inria ) -Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique ( CNRS ) -Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique ( Inria ) -Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique ( CNRS ) -Inria Lille - Nord Europe, and ANR-12-MONU-0009,NumBBO,Analyse, Amelioration, Evaluation d'algorithmes numériques pour l'optimisation boîte-noire ( 2012 )
Subjects: FOS: Computer and information sciences, [INFO.INFO-PF]Computer Science [cs]/Performance [cs.PF], [ INFO.INFO-MS ] Computer Science [cs]/Mathematical Software [cs.MS], [ INFO.INFO-NA ] Computer Science [cs]/Numerical Analysis [cs.NA], [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Computer Science - Neural and Evolutionary Computing, Neural and Evolutionary Computing (cs.NE), [ INFO.INFO-NE ] Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], [INFO.INFO-NA]Computer Science [cs]/Numerical Analysis [cs.NA], [ INFO.INFO-PF ] Computer Science [cs]/Performance [cs.PF], [ INFO.INFO-LG ] Computer Science [cs]/Machine Learning [cs.LG], [INFO.INFO-MS]Computer Science [cs]/Mathematical Software [cs.MS]
Abstract: ArXiv e-prints, arXiv:1605.03560; We present an any-time performance assessment for benchmarking numerical optimization algorithms in a black-box scenario, applied within the COCO benchmarking platform. The performance assessment is based on runtimes measured in number of objective function evaluations to reach one or several quality indicator target values. We argue that runtime is the only available measure with a generic, meaningful, and quantitative interpretation. We discuss the choice of the target values, runlength-based targets, and the aggregation of results by using simulated restarts, averages, and empirical distribution functions.
Published: 2016

7. Combination of One-Class Support Vector Machines for Classification with Reject Option

Author: Michèle Sebag, Blaise Hanczar, Laboratoire d'Informatique Paris Descartes (LIPADE - EA 2517), Université Paris Descartes - Paris 5 (UPD5), Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Calders, Toon and Esposito, Floriana and Hüllermeier, Eyke and Meo, Rosa, 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 d'Informatique Paris Descartes ( LIPADE - EA 2517 ), Université Paris Descartes - Paris 5 ( UPD5 ), Machine Learning and Optimisation ( TAO ), 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 ) -Université Paris-Sud - Paris 11 ( UP11 ) -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 ) -CentraleSupélec, 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 ), and Calders, Toon and Esposito, Floriana and Hüllermeier, Eyke and Meo, Rosa
Subjects: Support vector machines, business.industry, media_common.quotation_subject, Constrained optimization, Ambiguity, Rejection option, Machine learning, computer.software_genre, Rejection rate, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Support vector machine, Binary classification, Outlier, Supervised classification, Sequential minimal optimization, Abstaining clas-sifier, Data mining, Artificial intelligence, [ INFO.INFO-AI ] Computer Science [cs]/Artificial Intelligence [cs.AI], business, computer, Classifier (UML), media_common, Mathematics
Abstract: International audience; This paper focuses on binary classification with reject op-tion, enabling the classifier to detect and abstain hazardous decisions. While reject classification produces in more reliable decisions, there is a tradeoff between accuracy and rejection rate. Two type of rejection are considered: ambiguity and outlier rejection. The state of the art mostly handles ambiguity rejection and ignored outlier rejection. The proposed approach, referred as CONSUM, handles both ambiguity and outliers detection. Our method is based on a quadratic constrained optimization formulation, combining one-class support vector machines. An adapta-tion of the sequential minimal optimization algorithm is proposed to solve the minimization problem. The experimental study on both artifi-cial and real world datasets exams the sensitivity of the CONSUM with respect to the hyper-parameters and demonstrates the superiority of our approach.
Published: 2014
Full Text: View/download PDF

8. Multi-Objective AI Planning: Evaluating DAE-YAHSP on a Tunable Benchmark

Author: Pierre Savéant, Vincent Vidal, Johann Dreo, Marc Schoenauer, Mostepha Khouadjia, Machine Learning and Optimisation (TAO), 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), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), ONERA - The French Aerospace Lab [Toulouse], ONERA, Thales Research and Technology [Palaiseau], THALES [France], Robin C. Purshouse and Peter J. Fleming and Carlos M. Fonseca, 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, THALES, Machine Learning and Optimisation ( TAO ), 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 ) -Université Paris-Sud - Paris 11 ( UP11 ) -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 ) -CentraleSupélec, 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 ), and ONERA - The French Aerospace Lab ( Toulouse )
Subjects: FOS: Computer and information sciences, AI Planning, Computer Science - Artificial Intelligence, Computer science, business.industry, Contrast (statistics), 0102 computer and information sciences, 02 engineering and technology, 01 natural sciences, Multi-objective optimization, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Single objective, Artificial Intelligence (cs.AI), 010201 computation theory & mathematics, Automated planning and scheduling, 0202 electrical engineering, electronic engineering, information engineering, Benchmark (computing), 020201 artificial intelligence & image processing, Artificial intelligence, [ INFO.INFO-AI ] Computer Science [cs]/Artificial Intelligence [cs.AI], business
Abstract: All standard AI planners to-date can only handle a single objective, and the only way for them to take into account multiple objectives is by aggregation of the objectives. Furthermore, and in deep contrast with the single objective case, there exists no benchmark problems on which to test the algorithms for multi-objective planning. Divide and Evolve (DAE) is an evolutionary planner that won the (single-objective) deterministic temporal satisficing track in the last International Planning Competition. Even though it uses intensively the classical (and hence single-objective) planner YAHSP, it is possible to turn DAE-YAHSP into a multi-objective evolutionary planner. A tunable benchmark suite for multi-objective planning is first proposed, and the performances of several variants of multi-objective DAE-YAHSP are compared on different instances of this benchmark, hopefully paving the road to further multi-objective competitions in AI planning., 7th International Conference on Evolutionary Multi-Criterion Optimization (2013) To appearr. arXiv admin note: text overlap with arXiv:0804.3965 by other authors
Published: 2013

9. Efficient Distributed Monitoring with Active Collaborative Prediction

Author: Dawei Feng, Tristan Glatard, Cécile Germain, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Images et Modèles, Centre de Recherche en Acquisition et Traitement de l'Image pour la Santé (CREATIS), Université Jean Monnet [Saint-Étienne] (UJM)-Hospices Civils de Lyon (HCL)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Jean Monnet [Saint-Étienne] (UJM)-Hospices Civils de Lyon (HCL)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM), France-Grilles FUI TIMCO, European Project: 261323,EC:FP7:INFRA,FP7-INFRASTRUCTURES-2010-2,EGI-INSPIRE(2010), 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 ), Machine Learning and Optimisation ( TAO ), Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique ( Inria ), 2 - Images et Modèles, Centre de Recherche en Acquisition et Traitement de l'Image pour la Santé ( CREATIS ), Université Claude Bernard Lyon 1 ( UCBL ), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon ( INSA Lyon ), Université de Lyon-Institut National des Sciences Appliquées ( INSA ) -Institut National des Sciences Appliquées ( INSA ) -Hospices Civils de Lyon ( HCL ) -Université Jean Monnet [Saint-Étienne] ( UJM ) -Institut National de la Santé et de la Recherche Médicale ( INSERM ) -Centre National de la Recherche Scientifique ( CNRS ) -Université Claude Bernard Lyon 1 ( UCBL ), Université de Lyon-Institut National des Sciences Appliquées ( INSA ) -Institut National des Sciences Appliquées ( INSA ) -Hospices Civils de Lyon ( HCL ) -Université Jean Monnet [Saint-Étienne] ( UJM ) -Institut National de la Santé et de la Recherche Médicale ( INSERM ) -Centre National de la Recherche Scientifique ( CNRS ), European Project : 261323,EC:FP7:INFRA,FP7-INFRASTRUCTURES-2010-2,EGI-INSPIRE ( 2010 ), 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), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université Jean Monnet - Saint-Étienne (UJM)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)-Université Claude Bernard Lyon 1 (UCBL), and Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université Jean Monnet - Saint-Étienne (UJM)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)
Subjects: Computer Networks and Communications, Computer science, Active learning (machine learning), Distributed computing, 02 engineering and technology, Fault (power engineering), Grid, computer.software_genre, [ INFO.INFO-LG ] Computer Science [cs]/Machine Learning [cs.LG], Matrix decomposition, [ INFO.INFO-DC ] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Hardware and Architecture, Margin (machine learning), 020204 information systems, Active learning, 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, Software, Selection (genetic algorithm)
Abstract: Isolating users from the inevitable faults in large distributed systems is critical to Quality of Experience. We formulate the problem of probe selection for fault prediction based on end-to-end probing as a Collaborative Prediction (CP) problem. On an extensive experimental dataset from the EGI (European Grid Initiative) grid, the combination of the Maximum Margin Matrix Factorization approach to CP and Active Learning shows excellent performance, reducing the number of probes typically by 80% to 90%. Comparison with other Collaborative Prediction strategies show that Active Probing is most efficient at dealing with the various sources of data variability.
Published: 2013
Full Text: View/download PDF

10. Scalable structural break detection

Author: Cécile Germain-Renaud, Pascal Bondon, T. íLtetö, Nikolaus Hansen, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Laboratoire des signaux et systèmes (L2S), DIGITEO / Ile de France Regional council, Software and Complex Systems program, Machine Learning and Optimisation ( TAO ), 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 ) -Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique ( Inria ), Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ), Division Signaux - L2S, Laboratoire des signaux et systèmes ( L2S ), 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 ), 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, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: Mathematical optimization, 020206 networking & telecommunications, Statistical model, 02 engineering and technology, 01 natural sciences, [ INFO.INFO-LG ] Computer Science [cs]/Machine Learning [cs.LG], 010104 statistics & probability, Chromosome (genetic algorithm), Autoregressive model, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], 0202 electrical engineering, electronic engineering, information engineering, Piecewise, 0101 mathematics, CMA-ES, Minimum description length, Evolution strategy, Representation (mathematics), Algorithm, Software, Mathematics
Abstract: International audience; This paper deals with a statistical model fitting procedure for non-stationary time series. This procedure selects the parameters of a piecewise autoregressive model using the Minimum Description Length principle. The existing chromosome representation of the piecewise autoregressive model and its corresponding optimisation algorithm are improved. First, we show that our proposed chromosome representation better captures the intrinsic properties of the piecewise autoregressive model. Second, we apply an optimisation algorithm, the Covariance Matrix Adaptation - Evolution Strategy, with which our setup converges faster to the optimal fit. Our proposed method achieves at least one order of magnitude performance improvement compared to the existing solution.
Published: 2012
Full Text: View/download PDF

11. Towards Non-Stationary Grid Models

Author: Michèle Sebag, Cécile Germain-Renaud, Pascal Bondon, Tamás Éltető, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Laboratoire des signaux et systèmes (L2S), France-Grilles, 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), Machine Learning and Optimisation ( TAO ), 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 ) -Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique ( Inria ), 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 des signaux et systèmes ( L2S ), and Université Paris-Sud - Paris 11 ( UP11 ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS )
Subjects: Mathematical optimization, Computer Networks and Communications, Computer science, media_common.quotation_subject, Complex system, 02 engineering and technology, computer.software_genre, 01 natural sciences, 010104 statistics & probability, [ INFO.INFO-DC ] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], 0202 electrical engineering, electronic engineering, information engineering, Quality (business), 0101 mathematics, media_common, Sequence, Bootstrapping, Model selection, Grid, Range (mathematics), Autoregressive model, Hardware and Architecture, 020201 artificial intelligence & image processing, Data mining, [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], computer, Software, Information Systems
Abstract: Despite intense research on Grid scheduling, differentiated quality of service remains an open question, and no consensus has emerged on the most promising strategy. The difficulties of experimentation might be one of the root causes of this stalling. An alternative to experimenting on real, large, and complex data is to look for well-founded and parsimonious representations, which may also contribute to the a-priori knowledge required for operational Autonomics. The goal of this paper is thus to explore explanatory and generative models rather than predictive ones. As a test case, we address the following issue: is it possible to exhibit and validate consistent models of the Grid workload? Most existing work on modeling the dynamics of Grid behavior describes Grids as complex systems, but assumes a steady-state system (technically stationarity) and concludes to some form of long-range dependence (slowly decaying correlation) in the associated time-series. But the physical (economic and sociologic) processes governing the Grid behavior dispel the stationarity hypothesis. This paper considers an appealing different class of models: a sequence of stationary processes separated by breakpoints. The model selection question is now defined as identifying the breakpoints and fitting the processes in each segment. Experimenting with data from the EGEE/EGI Grid, we found that a non-stationary model can consistently be identified from empirical data, and that limiting the range of models to piecewise affine (autoregressive) time series is sufficiently powerful. We propose and experiment a validation methodology that empirically addresses the current lack of theoretical results concerning the quality of the estimated model parameters. Finally, we present a bootstrapping strategy for building more robust models from the limited samples at hand.
Published: 2011
Full Text: View/download PDF

12. Multi-objective reinforcement learning for responsive grids

Author: Cécile Germain-Renaud, Charles Loomis, Balázs Kégl, Julien Perez, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Laboratoire de l'Accélérateur Linéaire (LAL), Université Paris-Sud - Paris 11 (UP11)-Institut National de Physique Nucléaire et de Physique des Particules du CNRS (IN2P3)-Centre National de la Recherche Scientifique (CNRS), DIM-LSC DIGITEO, Grid Observatory, European Project: 222667,EC:FP7:INFRA,FP7-INFRASTRUCTURES-2007-2,EGEE-III(2008), 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 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ), Machine Learning and Optimisation ( TAO ), Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique ( Inria ), Laboratoire de l'Accélérateur Linéaire ( LAL ), Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Physique Nucléaire et de Physique des Particules du CNRS ( IN2P3 ) -Centre National de la Recherche Scientifique ( CNRS ), and European Project : 222667,EC:FP7:INFRA,FP7-INFRASTRUCTURES-2007-2,EGEE-III ( 2008 )
Subjects: [ INFO.INFO-MO ] Computer Science [cs]/Modeling and Simulation, Computer Networks and Communications, Computer science, Distributed computing, Cloud computing, 02 engineering and technology, [ INFO.INFO-LG ] Computer Science [cs]/Machine Learning [cs.LG], Autonomic computing, Resource (project management), [ INFO.INFO-DC ] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], 0202 electrical engineering, electronic engineering, information engineering, Reinforcement learning, business.industry, 020206 networking & telecommunications, Provisioning, Grid, [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation, Shared resource, Hardware and Architecture, 020201 artificial intelligence & image processing, Echo state network, [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], business, Software, Information Systems
Abstract: The original publication is available at www.springerlink.com; International audience; Grids organize resource sharing, a fundamental requirement of large scientific collaborations. Seamless integration of grids into everyday use requires responsiveness, which can be provided by elastic Clouds, in the Infrastructure as a Service (IaaS) paradigm. This paper proposes a model-free resource provisioning strategy supporting both requirements. Provisioning is modeled as a continuous action-state space, multi-objective reinforcement learning (RL) problem, under realistic hypotheses; simple utility functions capture the high level goals of users, administrators, and shareholders. The model-free approach falls under the general program of autonomic computing, where the incremental learning of the value function associated with the RL model provides the so-called feedback loop. The RL model includes an approximation of the value function through an Echo State Network. Experimental validation on a real data-set from the EGEE grid shows that introducing a moderate level of elasticity is critical to ensure a high level of user satisfaction.
Published: 2010
Full Text: View/download PDF

13. Grid Analysis of Radiological Data

Author: Germain-Renaud, Cecile, Breton, Vincent, Clarysse, Patrick, Delhay, Bertrand, Gaudeau, Yann, Glatard, Tristan, Jeannot, Emmanuel, Legre, Yannick, Montagnat, Johan, Moureaux, Jean-Marie, Osorio, Angel, Pennec, Xavier, Schaerer, Joel, Texier, Romain, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Machine Learning and Optimisation (TAO), 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), Equipe PCSV, Laboratoire de Physique Corpusculaire - Clermont-Ferrand (LPC), Université Blaise Pascal - Clermont-Ferrand 2 (UBP)-Institut National de Physique Nucléaire et de Physique des Particules du CNRS (IN2P3)-Centre National de la Recherche Scientifique (CNRS)-Université Blaise Pascal - Clermont-Ferrand 2 (UBP)-Institut National de Physique Nucléaire et de Physique des Particules du CNRS (IN2P3)-Centre National de la Recherche Scientifique (CNRS), Centre de Recherche et d'Application en Traitement de l'Image et du Signal (CREATIS), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-École Supérieure de Chimie Physique Électronique de Lyon (CPE)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS), Centre de Recherche en Automatique de Nancy (CRAN), Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), Algorithms for the Grid (ALGORILLE), INRIA Lorraine, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université Henri Poincaré - Nancy 1 (UHP)-Université Nancy 2-Institut National Polytechnique de Lorraine (INPL)-Centre National de la Recherche Scientifique (CNRS)-Université Henri Poincaré - Nancy 1 (UHP)-Université Nancy 2-Institut National Polytechnique de Lorraine (INPL)-Centre National de la Recherche Scientifique (CNRS), HealthGrid, Laboratoire d'Informatique, Signaux, et Systèmes de Sophia-Antipolis (I3S) / Equipe MODALIS, Scalable and Pervasive softwARe and Knowledge Systems (Laboratoire I3S - SPARKS), Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis (I3S), Université Nice Sophia Antipolis (1965 - 2019) (UNS), COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UCA)-Université Nice Sophia Antipolis (1965 - 2019) (UNS), COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UCA)-Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis (I3S), COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UCA), Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur (LIMSI), Université Paris-Sud - Paris 11 (UP11)-Sorbonne Université - UFR d'Ingénierie (UFR 919), Sorbonne Université (SU)-Sorbonne Université (SU)-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)-Université Paris Saclay (COmUE), Analysis and Simulation of Biomedical Images (ASCLEPIOS), Inria Sophia Antipolis - Méditerranée (CRISAM), The AGIR project has been funded by the French Ministry of Research under the Masses de Données program. Developments has been partially supported by the EGEE-II project funded by the European Union INFSO-RI-031688, by other grants from the European Commission (BioinfoGRID, Embrace), and for the work in section 7 by regional authorities (Conseil Régional d'Auvergne, Conseil Général du Puy-de-Dôme, Conseil Général de l'Allier), Mario Cannataro (Ed.), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-École Supérieure Chimie Physique Électronique de Lyon-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS), Centre National de la Recherche Scientifique (CNRS)-Institut National Polytechnique de Lorraine (INPL)-Université Nancy 2-Université Henri Poincaré - Nancy 1 (UHP)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Institut National Polytechnique de Lorraine (INPL)-Université Nancy 2-Université Henri Poincaré - Nancy 1 (UHP), Université Nice Sophia Antipolis (... - 2019) (UNS), COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UCA)-Université Nice Sophia Antipolis (... - 2019) (UNS), Université Paris Saclay (COmUE)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université - UFR d'Ingénierie (UFR 919), Sorbonne Université (SU)-Sorbonne Université (SU)-Université Paris-Saclay-Université Paris-Sud - Paris 11 (UP11), 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 ), Machine Learning and Optimisation ( TAO ), Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Université Paris-Sud - Paris 11 ( UP11 ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -CentraleSupélec-Centre National de la Recherche Scientifique ( CNRS ) -Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique ( Inria ), Laboratoire de Physique Corpusculaire - Clermont-Ferrand ( LPC ), Université Blaise Pascal - Clermont-Ferrand 2 ( UBP ) -Institut National de Physique Nucléaire et de Physique des Particules du CNRS ( IN2P3 ) -Centre National de la Recherche Scientifique ( CNRS ) -Université Blaise Pascal - Clermont-Ferrand 2 ( UBP ) -Institut National de Physique Nucléaire et de Physique des Particules du CNRS ( IN2P3 ) -Centre National de la Recherche Scientifique ( CNRS ), Centre de Recherche et d'Application en Traitement de l'Image et du Signal ( CREATIS ), Université Claude Bernard Lyon 1 ( UCBL ), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon ( INSA Lyon ), Université de Lyon-Institut National des Sciences Appliquées ( INSA ) -Institut National des Sciences Appliquées ( INSA ) -École Supérieure Chimie Physique Électronique de Lyon-Institut National de la Santé et de la Recherche Médicale ( INSERM ) -Centre National de la Recherche Scientifique ( CNRS ), Centre de Recherche en Automatique de Nancy ( CRAN ), Université de Lorraine ( UL ) -Centre National de la Recherche Scientifique ( CNRS ), Algorithms for the Grid ( ALGORILLE ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Institut National de Recherche en Informatique et en Automatique ( Inria ) -Laboratoire Lorrain de Recherche en Informatique et ses Applications ( LORIA ), Institut National de Recherche en Informatique et en Automatique ( Inria ) -Université Henri Poincaré - Nancy 1 ( UHP ) -Université Nancy 2-Institut National Polytechnique de Lorraine ( INPL ) -Centre National de la Recherche Scientifique ( CNRS ) -Université Henri Poincaré - Nancy 1 ( UHP ) -Université Nancy 2-Institut National Polytechnique de Lorraine ( INPL ) -Centre National de la Recherche Scientifique ( CNRS ), Scalable and Pervasive softwARe and Knowledge Systems ( SPARKS ), Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis ( I3S ), Université Nice Sophia Antipolis ( UNS ), Université Côte d'Azur ( UCA ) -Université Côte d'Azur ( UCA ) -Centre National de la Recherche Scientifique ( CNRS ) -Université Nice Sophia Antipolis ( UNS ), Université Côte d'Azur ( UCA ) -Université Côte d'Azur ( UCA ) -Centre National de la Recherche Scientifique ( CNRS ) -Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis ( I3S ), Université Côte d'Azur ( UCA ) -Université Côte d'Azur ( UCA ) -Centre National de la Recherche Scientifique ( CNRS ), Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur ( LIMSI ), Université Paris-Sud - Paris 11 ( UP11 ) -Centre National de la Recherche Scientifique ( CNRS ), Analysis and Simulation of Biomedical Images ( ASCLEPIOS ), 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 ), Université Paris-Sud - Paris 11 (UP11) - Institut National de Recherche en Informatique et en Automatique (Inria) - CentraleSupélec - Centre National de la Recherche Scientifique (CNRS), Université Paris-Sud - Paris 11 (UP11) - Institut National de Recherche en Informatique et en Automatique (Inria) - CentraleSupélec - Centre National de la Recherche Scientifique (CNRS) - Université Paris-Sud - Paris 11 (UP11) - Institut National de Recherche en Informatique et en Automatique (Inria) - CentraleSupélec - Centre National de la Recherche Scientifique (CNRS) - Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria), Université Blaise Pascal - Clermont-Ferrand 2 (UBP) - Institut National de Physique Nucléaire et de Physique des Particules du CNRS (IN2P3) - Centre National de la Recherche Scientifique (CNRS) - Université Blaise Pascal - Clermont-Ferrand 2 (UBP) - Institut National de Physique Nucléaire et de Physique des Particules du CNRS (IN2P3) - Centre National de la Recherche Scientifique (CNRS), Centre de recherche en applications et traitement de l'image pour la santé (CREATIS), Centre National de la Recherche Scientifique (CNRS) - Institut National de la Santé et de la Recherche Médicale (INSERM) - Hospices Civils de Lyon - Université Claude Bernard Lyon 1 (UCBL) - Institut National des Sciences Appliquées (INSA), Université de Lorraine (UL) - Centre National de la Recherche Scientifique (CNRS), Institut National de Recherche en Informatique et en Automatique (Inria) - Institut National de Recherche en Informatique et en Automatique (Inria) - Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria) - Université Henri Poincaré - Nancy 1 (UHP) - Université Nancy 2 - Institut National Polytechnique de Lorraine (INPL) - Centre National de la Recherche Scientifique (CNRS) - Université Henri Poincaré - Nancy 1 (UHP) - Université Nancy 2 - Institut National Polytechnique de Lorraine (INPL) - Centre National de la Recherche Scientifique (CNRS), Scalable and Pervasive softwARe and Knowledge Systems (SPARKS), Université Nice Sophia Antipolis (UNS) - Centre National de la Recherche Scientifique (CNRS) - Université Nice Sophia Antipolis (UNS) - Centre National de la Recherche Scientifique (CNRS) - Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis (I3S), Université Nice Sophia Antipolis (UNS) - Centre National de la Recherche Scientifique (CNRS) - Université Nice Sophia Antipolis (UNS) - Centre National de la Recherche Scientifique (CNRS), Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur [Orsay] (LIMSI), Université Pierre et Marie Curie - Paris 6 (UPMC) - Université Paris-Sud - Paris 11 (UP11) - Centre National de la Recherche Scientifique (CNRS), Institut National de Recherche en Informatique et en Automatique (Inria) - Institut National de Recherche en Informatique et en Automatique (Inria), and Montagnat, Johan
Subjects: 020203 distributed computing, Data processing, Telemedicine, Exploit, business.industry, Computer science, Data management, 02 engineering and technology, Grid, Data science, 030218 nuclear medicine & medical imaging, 03 medical and health sciences, 0302 clinical medicine, Semantic grid, Workflow, [ INFO.INFO-DC ] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], 0202 electrical engineering, electronic engineering, information engineering, [INFO.INFO-DC] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], Use case, [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], business
Abstract: IGI-Global Medical Information Science Discoveries Research Award 2009; International audience; Grid technologies and infrastructures can contribute to harnessing the full power of computer-aided image analysis into clinical research and practice. Given the volume of data, the sensitivity of medical information, and the joint complexity of medical datasets and computations expected in clinical practice, the challenge is to fill the gap between the grid middleware and the requirements of clinical applications. This chapter reports on the goals, achievements and lessons learned from the AGIR (Grid Analysis of Radiological Data) project. AGIR addresses this challenge through a combined approach. On one hand, leveraging the grid middleware through core grid medical services (data management, responsiveness, compression, and workflows) targets the requirements of medical data processing applications. On the other hand, grid-enabling a panel of applications ranging from algorithmic research to clinical use cases both exploits and drives the development of the services.
Published: 2009
Full Text: View/download PDF

14. COCO: A Platform for Comparing Continuous Optimizers in a Black-Box Setting

Author: Nikolaus Hansen, Dimo Brockhoff, Raymond Ros, Olaf Mersmann, Anne Auger, Tea Tušar, Randomized Optimisation (RANDOPT ), Centre de Mathématiques Appliquées - Ecole Polytechnique (CMAP), École polytechnique (X)-Centre National de la Recherche Scientifique (CNRS)-École polytechnique (X)-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), Jozef Stefan Institute [Ljubljana] (IJS), Technische Universität Dortmund [Dortmund] (TU), ANR-12-MONU-0009,NumBBO,Analyse, Amelioration, Evaluation d'algorithmes numériques pour l'optimisation boîte-noire(2012), Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Parallel Cooperative Multi-criteria Optimization (DOLPHIN), 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 (CRIStAL) - UMR 9189 (CRIStAL), Centre National de la Recherche Scientifique (CNRS)-Université de Lille-Ecole Centrale de Lille-Centre National de la Recherche Scientifique (CNRS)-Université de Lille-Ecole Centrale de Lille, and CentraleSupélec-Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec
Subjects: FOS: Computer and information sciences, Optimization, Control and Optimization, Computer Science - Artificial Intelligence, 0211 other engineering and technologies, Machine Learning (stat.ML), Numerical optimization, 02 engineering and technology, [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], computer.software_genre, Software, [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], Statistics - Machine Learning, Derivative-free optimization, 0202 electrical engineering, electronic engineering, information engineering, Code (cryptography), FOS: Mathematics, Relevance (information retrieval), Mathematics - Numerical Analysis, Mathematics, [INFO.INFO-MS]Computer Science [cs]/Mathematical Software [cs.MS], Black box (phreaking), Structure (mathematical logic), derivative-free optimization, Measure (data warehouse), 021103 operations research, runtime distributions, Programming language, business.industry, software, Applied Mathematics, test functions, Benchmarking, Numerical Analysis (math.NA), [INFO.INFO-NA]Computer Science [cs]/Numerical Analysis [cs.NA], [INFO.INFO-PF]Computer Science [cs]/Performance [cs.PF], Artificial Intelligence (cs.AI), Black-box optimization, Computer Science - Mathematical Software, 020201 artificial intelligence & image processing, performance assessment, business, computer, Mathematical Software (cs.MS)
Abstract: We introduce COCO, an open source platform for Comparing Continuous Optimizers in a black-box setting. COCO aims at automatizing the tedious and repetitive task of benchmarking numerical optimization algorithms to the greatest possible extent. The platform and the underlying methodology allow to benchmark in the same framework deterministic and stochastic solvers for both single and multiobjective optimization. We present the rationales behind the (decade-long) development of the platform as a general proposition for guidelines towards better benchmarking. We detail underlying fundamental concepts of COCO such as the definition of a problem as a function instance, the underlying idea of instances, the use of target values, and runtime defined by the number of function calls as the central performance measure. Finally, we give a quick overview of the basic code structure and the currently available test suites., Comment: Optimization Methods and Software, Taylor & Francis, In press, pp.1-31
Published: 2021
Full Text: View/download PDF

15. The non-stationary stochastic multi-armed bandit problem

Author: Raphaël Féraud, Robin Allesiardo, Odalric-Ambrym Maillard, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Orange Labs [Lannion], France Télécom, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, ANR-16-CE40-0002,BADASS,BANDITS MANCHOTS POUR SIGNAUX NON-STATIONNAIRES ET STRUCTURES(2016), 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, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: Independent and identically distributed random variables, Mathematical optimization, Sample complexity, Stochastic process, Applied Mathematics, Existential quantification, Regret, 02 engineering and technology, 01 natural sciences, Multi-armed bandit, Measure (mathematics), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Computer Science Applications, 010104 statistics & probability, Identification (information), Computational Theory and Mathematics, [MATH.MATH-ST]Mathematics [math]/Statistics [math.ST], Modeling and Simulation, 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, 0101 mathematics, Astrophysics::Galaxy Astrophysics, Information Systems, Mathematics
Abstract: International audience; We consider a variant of the stochastic multi-armed bandit with K arms where the rewards are not assumed to be identically distributed, but are generated by a non-stationary stochastic process. We first study the unique best arm setting when there exists one unique best arm. Second, we study the general switching best arm setting when a best arm switches at some unknown steps. For both settings, we target problem-dependent bounds, instead of the more conservative problem-free bounds. We consider two classical problems: (1) identify a best arm with high probability (best arm identification), for which the performance measure by the sample complexity (number of samples before finding a near-optimal arm). To this end, we naturally extend the definition of sample complexity so that it makes sense in the switching best arm setting, which may be of independent interest. (2) Achieve the smallest cumulative regret (regret minimization) where the regret is measured with respect to the strategy pulling an arm with the best instantaneous mean at each step.
Published: 2017
Full Text: View/download PDF

16. Convolutional Neural Networks for Large-Scale Remote-Sensing Image Classification

Author: Pierre Alliez, Emmanuel Maggiori, Yuliya Tarabalka, Guillaume Charpiat, Geometric Modeling of 3D Environments (TITANE), 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), Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Laboratoire de Recherche en Informatique (LRI), and 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
Subjects: 010504 meteorology & atmospheric sciences, Computer science, 0211 other engineering and technologies, Context (language use), 02 engineering and technology, computer.software_genre, 01 natural sciences, Convolutional neural network, convolutional neural networks, Relevance (information retrieval), Electrical and Electronic Engineering, ComputingMilieux_MISCELLANEOUS, 021101 geological & geomatics engineering, 0105 earth and related environmental sciences, Training set, Contextual image classification, business.industry, Deep learning, deep learning, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Pattern recognition, Classification, [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV], General Earth and Planetary Sciences, Data mining, Artificial intelligence, Scale (map), business, computer, satellite images
Abstract: We propose an end-to-end framework for the dense, pixelwise classification of satellite imagery with convolutional neural networks (CNNs). In our framework, CNNs are directly trained to produce classification maps out of the input images. We first devise a fully convolutional architecture and demonstrate its relevance to the dense classification problem. We then address the issue of imperfect training data through a two-step training approach: CNNs are first initialized by using a large amount of possibly inaccurate reference data, and then refined on a small amount of accurately labeled data. To complete our framework, we design a multiscale neuron module that alleviates the common tradeoff between recognition and precise localization. A series of experiments show that our networks consider a large amount of context to provide fine-grained classification maps.
Published: 2017
Full Text: View/download PDF

17. Accelerated Optimization Methods for Force‐Field Parametrization in Battery Electrode Manufacturing Modeling

Author: Alain C. Ngandjong, Jean-Baptiste Hoock, Emiliano N. Primo, Marc Duquesnoy, Alejandro A. Franco, Teo Lombardo, Laboratoire réactivité et chimie des solides - UMR CNRS 7314 (LRCS), Université de Picardie Jules Verne (UPJV)-Institut de Chimie du CNRS (INC)-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), Machine Learning and Optimisation (TAO), 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), Réseau sur le stockage électrochimique de l'énergie (RS2E), Université de Nantes (UN)-Aix Marseille Université (AMU)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Collège de France (CdF (institution))-Université de Picardie Jules Verne (UPJV)-Ecole Nationale Supérieure de Chimie de Montpellier (ENSCM)-Ecole Nationale Supérieure de Chimie de Paris - Chimie ParisTech-PSL (ENSCP), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Université de Pau et des Pays de l'Adour (UPPA)-Institut de Chimie du CNRS (INC)-Université de Montpellier (UM)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Institut National Polytechnique (Toulouse) (Toulouse INP), Université de Toulouse (UT)-Université de Toulouse (UT)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Université Grenoble Alpes (UGA), Advanced Lithium Energy Storage Systems - ALISTORE-ERI (ALISTORE-ERI), and Institut de Chimie du CNRS (INC)-Centre National de la Recherche Scientifique (CNRS)
Subjects: Materials science, Energy Engineering and Power Technology, Particle swarm optimization, Mechanical engineering, [CHIM.MATE]Chemical Sciences/Material chemistry, 02 engineering and technology, 010402 general chemistry, 021001 nanoscience & nanotechnology, 01 natural sciences, Force field (chemistry), 0104 chemical sciences, Battery electrode, Electrochemistry, Optimization methods, Electrical and Electronic Engineering, 0210 nano-technology
Abstract: International audience; The performance, durability and cost of modern Li-ion batteries (LIBs) strongly depend on their manufacturing process. In this regard, computational methods that attempt to model LIBs manufacturing have the potential to help experimentalists to reduce costs and improve performances of LIBs. To do so, the electrode slurry phase, consisting of a suspension of active material, carbon additive and binder in a solvent must be modeled at first at the mesoscopic scale. However, efforts in that sense are rare in literature due to the inherent complexity of slurries and the difficulty of setting up appropriate metrics for the validation of the modeling results. For the first time, we propose an approach relying on experimental data that allows to validate the properties of simulated Nickel-Manganese-Cobalt-based slurries with different compositions and solid contents through force fields (FFs) parametrization. The latter was attained by Coarse-Grained Molecular Dynamics (CGMD), which enables to model the slurry phase at the mesoscopic scale and to consider explicitly the carbon binder domain. Moreover, we demonstrate that the CGMD FFs parameterization can be accelerated thanks to different types of Particle Swarm Optimization-based algorithms, which would allow faster screening of different simulated slurries fabrication conditions.
Published: 2020
Full Text: View/download PDF

18. Model of Intention for the model specification verification during the simulation architecture design process

Author: Olivia Penas, Eric Duceau, Eric Landel, Romain Barbedienne, Mouadh Yagoubi, IRT SystemX (IRT SystemX), Laboratoire d'Ingénierie des Systèmes Mécaniques et des MAtériaux (LISMMA), Université Paris 8 Vincennes-Saint-Denis (UP8)-SUPMECA - Institut supérieur de mécanique de Paris (SUPMECA), Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Airbus [France], Développement des Technologies Nouvelles (DTN), École Nationale Supérieure de Techniques Avancées Bretagne (ENSTA Bretagne), and CentraleSupélec-Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec
Subjects: Standards, Microwave integrated circuits, Computer science, Process (engineering), Numerical models, 0211 other engineering and technologies, Specification model, 02 engineering and technology, Architecture design, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], [SPI]Engineering Sciences [physics], Mathematical model, 0202 electrical engineering, electronic engineering, information engineering, Context modeling, Computer architecture, Architecture, Context model, 021103 operations research, business.industry, Computational modeling, computer.file_format, [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation, Verification model, Specification, 020201 artificial intelligence & image processing, Executable, Model of intention, Engineering design process, Software engineering, business, computer, Realization (systems), Simulation architecture
Abstract: International audience; Nowadays, simulation technologies are becoming indispensable to support the design process, and notably during the verification and optimization phases. However, when building a simulation architecture that has to provide the systems architects with answers, simulations architects have to correctly specify the corresponding models required. In a MBSE concept, we propose to extend the concept of model of intention, to verify the specification of these models, to model suppliers by providing them with a Model Request Package. Models of Intention will allow an executable verification of the simulation architecture before the integration of the final models of realization.
Published: 2018
Full Text: View/download PDF

19. The RAMP framework: from reproducibility to transparency in the design and optimization of scientific workflows

Author: Kégl, Balázs, Boucaud, Alexandre, Cherti, Mehdi, Kazakci, Akin, Gramfort, Alexandre, Lemaitre, Guillaume, Van den Bossche, Joris, Benbouzid, Djalel, Marini, Camille, Machine Learning and Optimisation (TAO), 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 l'Accélérateur Linéaire (LAL), Université Paris-Sud - Paris 11 (UP11)-Institut National de Physique Nucléaire et de Physique des Particules du CNRS (IN2P3)-Centre National de la Recherche Scientifique (CNRS), Centre de Gestion Scientifique i3 (CGS i3), Mines Paris - PSL (École nationale supérieure des mines de Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut interdisciplinaire de l’innovation (I3), Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS), Modelling brain structure, function and variability based on high-field MRI data (PARIETAL), Service NEUROSPIN (NEUROSPIN), Université Paris-Saclay-Direction de Recherche Fondamentale (CEA) (DRF (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Paris-Saclay-Direction de Recherche Fondamentale (CEA) (DRF (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Inria Saclay - Ile de France, Owkin France, 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), CentraleSupélec-Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Centre National de la Recherche Scientifique (CNRS)-Institut National de Physique Nucléaire et de Physique des Particules du CNRS (IN2P3)-Université Paris-Sud - Paris 11 (UP11), MINES ParisTech - École nationale supérieure des mines de Paris, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-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)-Service NEUROSPIN (NEUROSPIN), Direction de Recherche Fondamentale (CEA) (DRF (CEA)), and Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Paris-Saclay
Subjects: [INFO]Computer Science [cs], ComputingMilieux_MISCELLANEOUS, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract: International audience
Published: 2018

20. Piecewise rigid curve deformation via a Finsler steepest descent

Author: Guillaume Charpiat, Giacomo Nardi, Gabriel Peyré, François-Xavier Vialard, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Spatio-Temporal Activity Recognition Systems (STARS), 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), CEntre de REcherches en MAthématiques de la DEcision (CEREMADE), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL), Méthodes numériques pour le problème de Monge-Kantorovich et Applications en sciences sociales (MOKAPLAN), Inria Paris-Rocquencourt, European Project: 279593,EC:FP7:ERC,ERC-2011-StG_20101014,SIGMA-VISION(2011), 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, Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS)
Subjects: Optimization problem, Banach space, Space (mathematics), 01 natural sciences, shape registration, gradient flow, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, FOS: Mathematics, Applied mathematics, Mathematics - Numerical Analysis, 0101 mathematics, Mathematics - Optimization and Control, ComputingMilieux_MISCELLANEOUS, Mathematics, Descent (mathematics), Applied Mathematics, Numerical Analysis (math.NA), Curve evolution, 010101 applied mathematics, Flow (mathematics), Optimization and Control (math.OC), Piecewise, [MATH.MATH-OC]Mathematics [math]/Optimization and Control [math.OC], Balanced flow, Gradient descent, [MATH.MATH-NA]Mathematics [math]/Numerical Analysis [math.NA], Finsler space
Abstract: This paper introduces a novel steepest descent flow in Banach spaces. This extends previous works on generalized gradient descent, notably the work of Charpiat et al. [15], to the setting of Finsler metrics. Such a generalized gradient allows one to take into account a prior on deformations (e.g., piecewise rigid) in order to favor some specific evolutions. We define a Finsler gradient descent method to minimize a functional defined on a Banach space and we prove a convergence theorem for such a method. In particular, we show that the use of non-Hilbertian norms on Banach spaces is useful to study non-convex optimization problems where the geometry of the space might play a crucial role to avoid poor local minima.We show some applications to the curve matching problem. In particular, we characterize piecewise rigid deformations on the space of curves and we study several models to perform piecewise rigid evolution of curves.
Published: 2016
Full Text: View/download PDF

21. Analysis of runtime of optimization algorithms for noisy functions over discrete codomains

Author: Youhei Akimoto, Olivier Teytaud, Sandra Astete-Morales, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, 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)-Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Inria Saclay - Ile de France, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: Mathematical optimization, General Computer Science, Logarithm, Gaussian, Function (mathematics), Theoretical Computer Science, symbols.namesake, Noise, Quadratic equation, Discrete Optimization, Factor (programming language), Discrete optimization, Resampling, Additive noise, symbols, Runtime analysis, [MATH.MATH-OC]Mathematics [math]/Optimization and Control [math.OC], computer, computer.programming_language, Mathematics
Abstract: We consider in this work the application of optimization algorithms to problems over discrete codomains corrupted by additive unbiased noise. We propose a modification of the algorithms by repeating the fitness evaluation of the noisy function sufficiently so that, with a fix probability, the function evaluation on the noisy case is identical to the true value. If the runtime of the algorithms on the noise-free case is known, the number of resampling is chosen accordingly. If not, the number of resampling is chosen regarding to the number of fitness evaluations, in an anytime manner. We conclude that if the additive noise is Gaussian, then the runtime on the noisy case, for an adapted algorithm using resamplings, is similar to the runtime on the noise-free case: we incur only an extra logarithmic factor. If the noise is non-Gaussian but with finite variance, then the total runtime of the noisy case is quadratic in function of the runtime on the noise-free case. (C) 2015 Elsevier B.V. All rights reserved., Article, THEORETICAL COMPUTER SCIENCE. 605:42-50 (2015)
Published: 2015
Full Text: View/download PDF

22. Latent binary MRF for online reconstruction of large scale systems

Author: Cyril Furtlehner, Jean-Marc Lasgouttes, Victorin Martin, Centre de Robotique (CAOR), Mines Paris - PSL (École nationale supérieure des mines de Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL), Robotics & Intelligent Transportation Systems (RITS), Inria Paris-Rocquencourt, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Machine Learning and Optimisation (TAO), 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, ANR TRAVESTI, ANR-08-SYSC-0017,TRAVESTI,Estimation du volume de Trafic par Inférence Spatio-temporelle(2008), MINES ParisTech - École nationale supérieure des mines de Paris, 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), and Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec
Subjects: inference, latent variables, Markov random field, Scale (ratio), soft constraints, Computer science, Applied Mathematics, Model selection, Inference, Latent variable, Belief propagation, 01 natural sciences, Mathematics Subject Classification (2010) 68T05 · 60K35, [MATH.MATH-PR]Mathematics [math]/Probability [math.PR], 010104 statistics & probability, 03 medical and health sciences, 0302 clinical medicine, Artificial Intelligence, Joint probability distribution, 0101 mathematics, Marginal distribution, Algorithm, 030217 neurology & neurosurgery, belief propagation
Abstract: International audience; We present a novel method for online inference of real-valued quantities on a large network from very sparse measurements. The target application is a large scale system, like e.g. a traffic network, where a small varying subset of the variables is observed, and predictions about the other variables have to be continuously updated. A key feature of our approach is the modeling of dependencies between the original variables through a latent binary Markov random field. This greatly simplifies both the model selection and its subsequent use. We introduce the mirror belief propagation algorithm, that performs fast inference in such a setting. The offline model estimation relies only on pairwise historical data and its complexity is linear w.r.t. the dataset size. Our method makes no assumptions about the joint and marginal distributions of the variables but is primarily designed with multimodal joint distributions in mind. Numerical experiments demonstrate both the applicability and scalability of the method in practice.
Published: 2015
Full Text: View/download PDF

23. Using Unlabeled Data to Discover Bivariate Causality with Deep Restricted Boltzmann Machines

Author: Nataliya Sokolovska, Jean-Daniel Zucker, Olga Permiakova, Sofia K. Forslund, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), CentraleSupélec-Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Sorbonne Université (SU), Laboratoire de Biologie à Grande Échelle (BGE - UMR S1038), Institut de Recherche Interdisciplinaire de Grenoble (IRIG), Direction de Recherche Fondamentale (CEA) (DRF (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Direction de Recherche Fondamentale (CEA) (DRF (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019]), European Molecular Biology Laboratory [Heidelberg] (EMBL), Unité de modélisation mathématique et informatique des systèmes complexes [Bondy] (UMMISCO), Institut de Recherche pour le Développement (IRD [France-Nord])-Institut de la francophonie pour l'informatique-Université Cheikh Anta Diop [Dakar, Sénégal] (UCAD)-Université Gaston Bergé (Saint-Louis, Sénégal)-Université Cadi Ayyad [Marrakech] (UCA)-Université de Yaoundé I-Sorbonne Université (SU), 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), Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Institut de Recherche Interdisciplinaire de Grenoble (IRIG), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA), Université de Yaoundé I-Institut de la francophonie pour l'informatique-Université Cheikh Anta Diop [Dakar, Sénégal] (UCAD)-Université Gaston Bergé (Saint-Louis, Sénégal)-Université Cadi Ayyad [Marrakech] (UCA)-Sorbonne Université (SU)-Institut de Recherche pour le Développement (IRD [France-Nord]), ANR-17-CE23-0006,DiagnoLearn,Apprentissage des modèles interprétables pour le diagnostic médical(2017), and Sokolovska, Nataliya
Subjects: semi-supervised learning, Computational complexity theory, 0206 medical engineering, 02 engineering and technology, Semi-supervised learning, Bivariate analysis, [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], Joint probability distribution, Databases, Genetic, metagenomic data, Genetics, Humans, Categorical variable, Models, Statistical, probabilistic models, Applied Mathematics, Computational Biology, Conditional probability distribution, [STAT.ML] Statistics [stat]/Machine Learning [stat.ML], Metformin, Gastrointestinal Microbiome, Causality, Causal inference, Metagenome, Neural Networks, Computer, Supervised Machine Learning, Marginal distribution, Algorithm, 020602 bioinformatics, Biotechnology
Abstract: An important question in microbiology is whether treatment causes changes in gut flora, and whether it also affects metabolism. The reconstruction of causal relations purely from non-temporal observational data is challenging. We address the problem of causal inference in a bivariate case, where the joint distribution of two variables is observed. We consider, in particular, data on discrete domains. The state-of-the-art causal inference methods for continuous data suffer from high computational complexity. Some modern approaches are not suitable for categorical data, and others need to estimate and fix multiple hyper-parameters. In this contribution, we introduce a novel method of causal inference which is based on the widely used assumption that if $X$ X causes $Y$ Y , then $P(X)$ P ( X ) and $P(Y|X)$ P ( Y | X ) are independent. We propose to explore a semi-supervised approach where $P(Y|X)$ P ( Y | X ) and $P(X)$ P ( X ) are estimated from labeled and unlabeled data respectively, whereas the marginal probability is estimated potentially from much more (cheap unlabeled) data than the conditional distribution. We validate the proposed method on the standard cause-effect pairs. We illustrate by experiments on several benchmarks of biological network reconstruction that the proposed approach is very competitive in terms of computational time and accuracy compared to the state-of-the-art methods. Finally, we apply the proposed method to an original medical task where we study whether drugs confound human metagenome.
Published: 2018
Full Text: View/download PDF

24. High-resolution image classification with convolutional networks

Author: Yuliya Tarabalka, Emmanuel Maggiori, Guillaume Charpiat, Pierre Alliez, Geometric Modeling of 3D Environments (TITANE), 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), Machine Learning and Optimisation (TAO), 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, 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, and Tarabalka, Yuliya
Subjects: [INFO.INFO-TS] Computer Science [cs]/Signal and Image Processing, Computer science, media_common.quotation_subject, Feature extraction, 0211 other engineering and technologies, 02 engineering and technology, Convolutional neural network, Image (mathematics), Convolution, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, convolutional neural networks, 0202 electrical engineering, electronic engineering, information engineering, Function (engineering), Image resolution, 021101 geological & geomatics engineering, media_common, business.industry, Deep learning, deep learning, 020207 software engineering, Pattern recognition, High-resolution aerial images, classification, High resolution image, Artificial intelligence, business
Abstract: International audience; We address the pixelwise classification of high-resolution aerial imagery. While convolutional neural networks (CNNs) are gaining increasing attention in image analysis, it is still challenging to adapt them to produce fine-grained classification maps. This is due to a well-known trade-off between recognition and localization: the impressive capability of CNNs to recognize meaningful objects comes at the price of losing spatial precision. We here propose an architecture that addresses this issue. It learns features at different levels of detail and also learns a function to combine them. By integrating local and global information in an efficient and flexible manner, it outperforms previous techniques.
Published: 2017

25. CMA-ES and advanced adaptation mechanisms

Author: Youhei Akimoto, Nikolaus Hansen, Machine Learning and Optimisation (TAO), 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), Randomized Optimisation (RANDOPT ), Centre de Mathématiques Appliquées - Ecole Polytechnique (CMAP), École polytechnique (X)-Centre National de la Recherche Scientifique (CNRS)-École polytechnique (X)-Centre National de la Recherche Scientifique (CNRS)-Inria Saclay - Ile de France, 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), and Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec
Subjects: [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], 010201 computation theory & mathematics, [INFO.INFO-CE]Computer Science [cs]/Computational Engineering, Finance, and Science [cs.CE], 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, 0102 computer and information sciences, 02 engineering and technology, [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], [INFO.INFO-NA]Computer Science [cs]/Numerical Analysis [cs.NA], 01 natural sciences, ComputingMilieux_MISCELLANEOUS
Abstract: International audience
Published: 2017
Full Text: View/download PDF

26. Editorial:special issue on computational intelligence for vision and robotics

Author: Alexandra Psarrou, Sergio Escalera, Jose Garcia-Rodriguez, Isabelle Guyon, Andrew Lewis, Miguel Cazorla, Universidad de Alicante, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Chalearn, Computer Vision Center (Centre de visio per computador) (CVC), Universitat Autònoma de Barcelona (UAB), University of Westminster [London] (UOW), Griffith University [Brisbane], Department of Software and Computing Systems [Alicante] (DLSI), 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, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: 0209 industrial biotechnology, Engineering, medicine.medical_specialty, Future of robotics, Management science, business.industry, Geography of robotics, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Computational intelligence, Robotics, 02 engineering and technology, 020901 industrial engineering & automation, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Artificial Intelligence, 0202 electrical engineering, electronic engineering, information engineering, medicine, Computational Science and Engineering, 020201 artificial intelligence & image processing, Artificial intelligence, business, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Software, ComputingMilieux_MISCELLANEOUS
Abstract: International audience; no abstract
Published: 2017
Full Text: View/download PDF

27. Enjeux sociaux des Big Data

Author: Tubaro, Paola, Casilli, Antonio, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Télécom ParisTech, Centre Edgar-Morin (CEM), Institut interdisciplinaire d'anthropologie du contemporain (IIAC), École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS), Mokrane Bouzeghoub, Rémy Mosseri, 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, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: données numériques, digital labor, partage des données, [SHS.SOCIO]Humanities and Social Sciences/Sociology, big data, [INFO]Computer Science [cs], vie privée
Abstract: National audience; Si la montée en puissance des "big data" est porteuse de promesses d'avancées scientifiques, de progrès dans la maîtrise des épidémies, d'amélioration du trafic urbain, et de nouvelles opportunités entrepreneuriales, il existe des risques sociaux qu'il convient de discuter. Le chapitre explore trois familles de risques, ceux qui touchent à la vie privée des citoyens et aux menaces accrues de surveillance, ceux qui ont trait aux conditions d'extraction et captation de valeur, et ceux qui découlent de l'opacité actuelle des droits de propriété et des possibilité de partage des données. Le chapitre conclut sur la nécessité de réfléchir à une régulation adaptée pour que les big data rapportent à l’usager plus qu’elles ne lui coûtent, sans pour autant entraver les avancées techniques et les opportunités économiques en résultant
Published: 2017

28. Explanatory Reasoning for Image Understanding Using Formal Concept Analysis and Description Logics

Author: Céline Hudelot, Jamal Atif, Isabelle Bloch, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Mathématiques Appliquées aux Systèmes - EA 4037 (MAS), Ecole Centrale Paris, Laboratoire Traitement et Communication de l'Information (LTCI), Télécom ParisTech-Institut Mines-Télécom [Paris] (IMT)-Centre National de la Recherche Scientifique (CNRS), 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, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: Theoretical computer science, Deductive reasoning, Description logics, image understanding, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Rationality, Model-based reasoning, Abductive reasoning, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Computer Science Applications, Human-Computer Interaction, formal concept analysis, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, Description logic, Control and Systems Engineering, explanatory reasoning, Formal concept analysis, mathematical morphology, Automated reasoning, Electrical and Electronic Engineering, Non-monotonic logic, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Software, Mathematics
Abstract: International audience; In this paper, we propose an original way of enriching description logics with abduction reasoning services. Under the aegis of set and lattice theories, we put together ingredients from mathematical morphology, description logics, and formal concept analysis. We propose computing the best explanations of an observation through algebraic erosion over the concept lattice of a background theory that is efficiently constructed using tools from formal concept analysis. We show that the defined operators are sound and complete and satisfy important rationality postulates of abductive reasoning. As a typical illustration, we consider a scene understanding problem. In fact, scene understanding can benefit from prior structural knowledge represented as an ontology and the reasoning tools of description logics. We formulate model based scene understanding as an abductive reasoning process. A scene is viewed as an observation and the interpretation is defined as the best explanation, considering the terminological knowledge part of a description logic about the scene context. This explanation is obtained from morphological operators applied on the corresponding concept lattice.
Published: 2014
Full Text: View/download PDF

29. Autoregressive models for gene regulatory network inference: Sparsity, stability and causality issues

Author: George Michailidis, Florence d'Alché-Buc, Department of Statistics, University of Michigan [Ann Arbor], University of Michigan System-University of Michigan System, Informatique, Biologie Intégrative et Systèmes Complexes (IBISC), Université d'Évry-Val-d'Essonne (UEVE), Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, 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), and GM was supported by NSF DMS-1228164 and NSA H98230-13-1-024 and FAB by the French National Research Agency [ANR-09-SYSC-009-01].
Subjects: Statistics and Probability, Gene regulatory network, Machine learning, computer.software_genre, 01 natural sciences, General Biochemistry, Genetics and Molecular Biology, 010104 statistics & probability, 03 medical and health sciences, Gene regulatory network inference, Econometrics, Cluster Analysis, Humans, Gene Regulatory Networks, 0101 mathematics, Cluster analysis, Dynamic Bayesian network, 030304 developmental biology, Mathematics, 0303 health sciences, Models, Genetic, General Immunology and Microbiology, business.industry, Applied Mathematics, Bayes Theorem, Genomics, General Medicine, Genome project, [SDV.BIBS]Life Sciences [q-bio]/Quantitative Methods [q-bio.QM], Quantitative Biology::Genomics, Causality, Dynamic Bayesian networks, Autoregressive model, Autoregressive models, Modeling and Simulation, Ordinary differential equation, Informatics, Artificial intelligence, [INFO.INFO-BI]Computer Science [cs]/Bioinformatics [q-bio.QM], General Agricultural and Biological Sciences, business, Stability, Sparsity, computer, Functional genomics
Abstract: International audience; Reconstructing gene regulatory networks from high-throughput measurements represents a key problem in functional genomics. It also represents a canonical learning problem and thus has attracted a lot of attention in both the informatics and the statistical learning literature. Numerous approaches have been proposed, ranging from simple clustering to rather involved dynamic Bayesian network modeling, as well as hybrid ones that combine a number of modeling steps, such as employing ordinary differential equations coupled with genome annotation. These approaches are tailored to the type of data being employed. Available data sources include static steady state data and time course data obtained either for wild type phenotypes or from perturbation experiments. This review focuses on the class of autoregressive models using time course data for inferring gene regulatory networks. The central themes of sparsity, stability and causality are discussed as well as the ability to integrate prior knowledge for successful use of these models for the learning task at hand.
Published: 2013
Full Text: View/download PDF

30. Tune and mix: learning to rank using ensembles of calibrated multi-class classifiers

Author: Róbert Busa-Fekete, György Szarvas, Tamás Éltető, Balázs Kégl, Laboratoire de l'Accélérateur Linéaire (LAL), Université Paris-Sud - Paris 11 (UP11)-Institut National de Physique Nucléaire et de Physique des Particules du CNRS (IN2P3)-Centre National de la Recherche Scientifique (CNRS), MTA-SZTE Research Group on Artificial Intelligence, University of Szeged [Szeged]-Hungarian Academy of Sciences (MTA), Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Ericsson [Budapest], Nuance Communications [Germany], 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, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: Ensemble methods, 02 engineering and technology, Machine learning, computer.software_genre, Regression Based Calibration, Ranking (information retrieval), Set (abstract data type), [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], Artificial Intelligence, Multi-class classification, 0202 electrical engineering, electronic engineering, information engineering, Relevance (information retrieval), Mathematics, Class Probability Calibration, business.industry, 05 social sciences, Probabilistic logic, Learning-to-rank, [INFO.INFO-RO]Computer Science [cs]/Operations Research [cs.RO], Ensemble learning, Support vector machine, [INFO.INFO-PF]Computer Science [cs]/Performance [cs.PF], ComputingMethodologies_PATTERNRECOGNITION, 020201 artificial intelligence & image processing, Learning to rank, Pairwise comparison, Artificial intelligence, Data mining, 0509 other social sciences, 050904 information & library sciences, business, computer, Software
Abstract: ANR-2010-COSI-002; In subset ranking, the goal is to learn a ranking function that approximates a gold standard partial ordering of a set of objects (in our case, a set of documents retrieved for the same query). The partial ordering is given by relevance labels representing the relevance of documents with respect to the query on an absolute scale. Our approach consists of three simple steps. First, we train standard multi-class classifiers (AdaBoost.MH and multi-class SVM) to discriminate between the relevance labels. Second, the posteriors of multi-class classifiers are calibrated using probabilistic and regression losses in order to estimate the Bayes-scoring function which optimizes the Normalized Discounted Cumulative Gain (NDCG). In the third step, instead of selecting the best multi-class hyperparameters and the best calibration, we mix all the learned models in a simple ensemble scheme. Our extensive experimental study is itself a substantial contribution. We compare most of the existing learning-to-rank techniques on all of the available large-scale benchmark data sets using a standardized implementation of the NDCG score. We show that our approach is competitive with conceptually more complex listwise and pairwise methods, and clearly outperforms them as the data size grows. As a technical contribution, we clarify some of the confusing results related to the ambiguities of the evaluation tools, and propose guidelines for future studies.
Published: 2013
Full Text: View/download PDF

31. Multidisciplinary Optimization in the Design of Future Space Launchers

Author: Isabelle Quinquis, Nathalie Delattre, Marc Schoenauer, Nikolaus Hansen, Guillaume Collange, Astrium Space Transportation (AST), ASTRIUM, Machine Learning and Optimisation (TAO), 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), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Rajan Filomeno Coelho and Piotr Breitkopf, 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, and Hansen, Nikolaus
Subjects: 020301 aerospace & aeronautics, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], 0203 mechanical engineering, Multidisciplinary approach, Computer science, 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Control engineering, [INFO.INFO-LG] Computer Science [cs]/Machine Learning [cs.LG], 02 engineering and technology, Space (mathematics)
Abstract: see https://www.lri.fr/~hansen/collange2010Chap12.pdf; International audience
Published: 2013
Full Text: View/download PDF

32. Quality Gain Analysis of the Weighted Recombination Evolution Strategy on General Convex Quadratic Functions

Author: Nikolaus Hansen, Youhei Akimoto, Anne Auger, Shinshu University [Nagano], Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Randomized Optimisation (RANDOPT ), Centre de Mathématiques Appliquées - Ecole Polytechnique (CMAP), École polytechnique (X)-Centre National de la Recherche Scientifique (CNRS)-École polytechnique (X)-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), Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria), Laboratoire de Recherche en Informatique (LRI), and 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
Subjects: Hessian matrix, general convex quadratic function, weighted recombination, General Computer Science, media_common.quotation_subject, 0102 computer and information sciences, 02 engineering and technology, recombination weights, [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], Space (mathematics), 01 natural sciences, Theoretical Computer Science, symbols.namesake, Quality (physics), Dimension (vector space), [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Search algorithm, FOS: Mathematics, 0202 electrical engineering, electronic engineering, information engineering, Applied mathematics, Limit (mathematics), quality gain analysis, Mathematics - Optimization and Control, media_common, Mathematics, optimal step-size, Regular polygon, Function (mathematics), Quadratic function, •Theory of computation → Theory of randomized search heuristics, [INFO.INFO-NA]Computer Science [cs]/Numerical Analysis [cs.NA], Infinity, Optimization and Control (math.OC), 010201 computation theory & mathematics, Evolution strategy, symbols, 020201 artificial intelligence & image processing, Keywords Evolution strategies, CCS Concepts •Mathematics of computing → Bio-inspired optimization
Abstract: Quality gain is the expected relative improvement of the function value in a single step of a search algorithm. Quality gain analysis reveals the dependencies of the quality gain on the parameters of a search algorithm, based on which one can derive the optimal values for the parameters. In this paper, we investigate evolution strategies with weighted recombination on general convex quadratic functions. We derive a bound for the quality gain and two limit expressions of the quality gain. From the limit expressions, we derive the optimal recombination weights and the optimal step-size, and find that the optimal recombination weights are independent of the Hessian of the objective function. Moreover, the dependencies of the optimal parameters on the dimension and the population size are revealed. Differently from previous works where the population size is implicitly assumed to be smaller than the dimension, our results cover the population size proportional to or greater than the dimension. Numerical simulation shows that the asymptotically optimal step-size well approximates the empirically optimal step-size for a finite dimensional convex quadratic function., Comment: Extended version of the work presented in FOGA 2017, accepted in TCS-D
Published: 2017
Full Text: View/download PDF

33. Algorithm Selector and Prescheduler in the ICON challenge

Author: Gonard, François, Schoenauer, Marc, Sebag, Michèle, IRT SystemX (IRT SystemX), TAckling the Underspecified (TAU), 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), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), El-Ghazali Talbi, Amir Nakib, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Investissements d'Avenir, and IRT SystemX
Subjects: Algorithm selection, Combinatorial optimization, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], [MATH.MATH-CO]Mathematics [math]/Combinatorics [math.CO], Empirical performance comparison, Performance prediction model, Algorithm portfolios, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract: International audience; Algorithm portfolios are known to offer robust performances, efficiently overcoming the weakness of every single algorithm on some particular problem instances. Two complementary approaches to get the best out of an algorithm portfolio is to achieve algorithm selection (AS), and to define a scheduler, sequentially launching a few algorithms on a limited computational budget each. The presented Algorithm Selector And Prescheduler system relies on the joint optimization of a pre-scheduler and a per instance AS, selecting an algorithm well-suited to the problem instance at hand. ASAP has been thoroughly evaluated against the state-of-the-art during the ICON challenge for algorithm selection, receiving an honourable mention. Its evaluation on several combinatorial optimization benchmarks exposes surprisingly good results of the simple heuristics used; some extensions thereof are presented and discussed in the paper.
Published: 2017

34. Multidimensional Decision Model for Classifying Learners: the case of Massive Online Open Courses (MOOCs)

Author: Brigui-Chtioui, Imène, Caillou, Philippe, International Business School [Paris] (ISG), Machine Learning and Optimisation (TAO), 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), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, emlyon business school, and business school, emlyon
Subjects: Multicriteria Decision model, Multiagent Theory, multi-criteria decision model, [SHS.GESTION]Humanities and Social Sciences/Business administration, MOOC, [SHS.ECO] Humanities and Social Sciences/Economics and Finance, [SHS.GESTION] Humanities and Social Sciences/Business administration, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, knowledge transfer, multi-agent theory, Clustering, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract: International audience; A new era of learning is arising, due to the development of the digital world and the maturity of web technologies. Massive open online courses (MOOCs) have emerged in this context and are challenging classical learning in spite of boundaries of time and space. They aim to provide good-quality education to masses that cannot be part of traditional university and school learning processes. This emancipation is coupled with a large amount of data collection via learning platforms. These data constitute a great opportunity to study interactions and to profit from this, in order to optimise learning and knowledge transfer. The challenge in creating value is how to represent, analyse and reuse data in order to characterise learners and so lead to better knowledge transfer. We focus, in this article, on the context of MOOC platforms based on Open edX (The leading online courses platform, initially developed by MIT and Harvard). The purpose of this paper is to propose a decision model that takes advantage of the different data-sets available on the learning platform in order to classify learners based on their behaviour and expertise. To this end, we propose a multi-agent approach that represents human actors using intelligent agents. A coordinator agent implements the multi-criteria decision model in order to optimise the learning process. This agent employs clustering algorithms and has an overview of the learning platform that enables it to assist learners and to learn from experiences and about other agents’ behaviour.
Published: 2017
Full Text: View/download PDF

35. A simple-to-use BDI architecture for agent-based modeling and simulation

Author: Benoit Gaudou, Patrick Taillandier, Chi Quang Truong, Arnaud Grignard, Philippe Caillou, Identités et Différenciation de l'Environnement des Espaces et des Sociétés (IDEES), Université de Rouen Normandie (UNIROUEN), Normandie Université (NU)-Normandie Université (NU)-Université de Caen Normandie (UNICAEN), Normandie Université (NU)-Université Le Havre Normandie (ULH), Normandie Université (NU)-Institut de Recherche Interdisciplinaire Homme et Société (IRIHS), Normandie Université (NU)-Normandie Université (NU)-Université de Rouen Normandie (UNIROUEN), Normandie Université (NU)-Centre National de la Recherche Scientifique (CNRS), Machine Learning and Optimisation (TAO), 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), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Systèmes Multi-Agents Coopératifs (IRIT-SMAC), Institut de recherche en informatique de Toulouse (IRIT), Université Toulouse Capitole (UT Capitole), Université de Toulouse (UT)-Université de Toulouse (UT)-Université Toulouse - Jean Jaurès (UT2J), Université de Toulouse (UT)-Université Toulouse III - Paul Sabatier (UT3), Université de Toulouse (UT)-Centre National de la Recherche Scientifique (CNRS)-Institut National Polytechnique (Toulouse) (Toulouse INP), Université de Toulouse (UT)-Toulouse Mind & Brain Institut (TMBI), Université Toulouse - Jean Jaurès (UT2J), Université de Toulouse (UT)-Université de Toulouse (UT)-Université Toulouse III - Paul Sabatier (UT3), Université de Toulouse (UT)-Université Toulouse Capitole (UT Capitole), Université de Toulouse (UT), Identité et Différenciation de l’Espace, de l’Environnement et des Sociétés (IDEES), Université de Caen Normandie (UNICAEN), Normandie Université (NU)-Normandie Université (NU)-Université Le Havre Normandie (ULH), Normandie Université (NU)-Université de Rouen Normandie (UNIROUEN), Normandie Université (NU)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche Interdisciplinaire Homme et Société (IRIHS), Normandie Université (NU), Unité de modélisation mathématique et informatique des systèmes complexes [Bondy] (UMMISCO), Institut de Recherche pour le Développement (IRD)-Université Pierre et Marie Curie - Paris 6 (UPMC)-Université de Yaoundé I-Institut de la francophonie pour l'informatique-Université Cheikh Anta Diop [Dakar, Sénégal] (UCAD)-Université Gaston Bergé (Saint-Louis, Sénégal)-Université Cadi Ayyad [Marrakech] (UCA), DREAM, DREAM-CTU/IRD, CICT, Can Tho University, Vietnam (DREAM), Can Tho University [Vietnam] (CTU)-Can Tho University [Vietnam] (CTU), ANR-14-CE22-0002,ACTEUR,Agents Cognitifs Territorialisés pour l'Etude des dynamiques Urbaines et des Risques(2014), Centre National de la Recherche Scientifique - CNRS (FRANCE), Institut National Polytechnique de Toulouse - INPT (FRANCE), Université de Rouen - UR (FRANCE), Université Pierre et Marie Curie, Paris 6 - UPMC (FRANCE), Université Toulouse III - Paul Sabatier - UT3 (FRANCE), Université Toulouse - Jean Jaurès - UT2J (FRANCE), Université Toulouse 1 Capitole - UT1 (FRANCE), Can Tho University - CTU ( VIETNAM), Université Paris-Sud 11 (FRANCE), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Université Toulouse 1 Capitole (UT1), Université Fédérale Toulouse Midi-Pyrénées-Université Fédérale Toulouse Midi-Pyrénées-Université Toulouse - Jean Jaurès (UT2J)-Université Toulouse III - Paul Sabatier (UT3), Université Fédérale Toulouse Midi-Pyrénées-Centre National de la Recherche Scientifique (CNRS)-Institut National Polytechnique (Toulouse) (Toulouse INP), Université Fédérale Toulouse Midi-Pyrénées-Université Toulouse 1 Capitole (UT1), Université Fédérale Toulouse Midi-Pyrénées, Université Cadi Ayyad [Marrakech] (UCA)-Université de Yaoundé I-Université Gaston Bergé (Saint-Louis, Sénégal)-Université Cheikh Anta Diop [Dakar, Sénégal] (UCAD)-Institut de la francophonie pour l'informatique-Université Pierre et Marie Curie - Paris 6 (UPMC), and Institut National Polytechnique de Toulouse - Toulouse INP (FRANCE)
Subjects: Cognitive model, Engineering, Modeling language, Distributed computing, [SDV]Life Sciences [q-bio], 02 engineering and technology, Modeling and simulation, 020204 information systems, 0202 electrical engineering, electronic engineering, information engineering, Reference architecture, Architecture, H- INFORMATIQUE, SIMPLE (military communications protocol), business.industry, Simulation framework, Modélisation et simulation, Service-oriented modeling, Multi Agent Simulation and Modelling, BDI architecture, BDI agent, Simulation Framework, [INFO.INFO-MA]Computer Science [cs]/Multiagent Systems [cs.MA], Systems engineering, 020201 artificial intelligence & image processing, Mekong delta, business
Abstract: International audience; With the increase of computing power and the development of user-friendly multi-agent simulation frameworks, social simulations have become increasingly realistic. However, most agent architectures in these simulations use simple reactive models. Cognitive architectures face two main obstacles: their complexity for the field-expert modeler, and their computational cost. In this paper , we propose a new cognitive agent architecture based on the BDI (Belief-Desire-Intention) paradigm integrated into the GAMA modeling platform. Based on the GAML modeling language, this architecture was designed to be simple-to-use for modelers, flexible enough to manage complex behaviors, and with low computational cost. This architecture is illustrated with a simulation of the evolution of land use in the Mekong Delta.
Published: 2017
Full Text: View/download PDF

36. ChaLearn Joint Contest on Multimedia Challenges Beyond Visual Analysis: An overview

Author: Isabelle Guyon, Martha Larson, Albert Clapés, Víctor Ponce-López, Henning Müller, Hugo Jair Escalante, Pål Halvorsen, Michael Riegler, Xavier Baró, Baiyu Chen, Jun Wan, Sergio Escalera, Instituto Nacional de Astrofísica, Óptica y Electrónica (INAOE), Computer Vision Center (Centre de visio per computador) (CVC), Universitat Autònoma de Barcelona (UAB), National Laboratory of Pattern Recognition [Beijing] (NLPR), Institute of Automation - Chinese Academy of Sciences, Simula Research Laboratory [Lysaker] (SRL), University of California [Berkeley], University of California, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Chalearn, Université Paris-Sud - Paris 11 - Faculté des Sciences (UP11 UFR Sciences), Université Paris-Sud - Paris 11 (UP11), University of Oslo (UiO), Haute Ecole Spécialisée de Suisse Occidentale (HES-SO), Delft University of Technology (TU Delft), University of California [Berkeley] (UC Berkeley), University of California (UC), 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, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: Computer science, 02 engineering and technology, Track (rail transport), CONTEST, computer.software_genre, User experience design, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], 0202 electrical engineering, electronic engineering, information engineering, CLIPS, Multimedia communication, computer.programming_language, Visualization, Benchmark testing, Organizations, Multimedia, business.industry, Deep learning, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], 020207 software engineering, Coopetition, Streaming media, 020201 artificial intelligence & image processing, Computer vision, State (computer science), Artificial intelligence, business, computer, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Protocols
Abstract: International audience; This paper provides an overview of the Joint Contest on Multimedia Challenges Beyond Visual Analysis. We organized an academic competition that focused on four problems that require effective processing of multimodal information in order to be solved. Two tracks were devoted to gesture spotting and recognition from RGB-D video, two fundamental problems for human computer interaction. Another track was devoted to a second round of the first impressions challenge of which the goal was to develop methods to recognize personality traits from short video clips. For this second round we adopted a novel collaborative-competitive (i.e., coopetition) setting. The fourth track was dedicated to the problem of video recommendation for improving user experience. The challenge was open for about 45 days, and received outstanding participation: almost 200 participants registered to the contest, and 20 teams sent predictions in the final stage. The main goals of the challenge were fulfilled: the state of the art was advanced considerably in the four tracks, with novel solutions to the proposed problems (mostly relying on deep learning). However, further research is still required. The data of the four tracks will be available to allow researchers to keep making progress in the four tracks.
Published: 2016
Full Text: View/download PDF

37. Random Shuffling and Resets for the Non-stationary Stochastic Bandit Problem

Author: Allesiardo, Robin, F��raud, Rapha��l, Maillard, Odalric-Ambrym, Machine Learning and Optimisation (TAO), 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), Orange Labs [Lannion], France Télécom, Laboratoire d'Informatique Fondamentale de Lille (LIFL), Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), and Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec
Subjects: FOS: Computer and information sciences, Artificial Intelligence (cs.AI), Computer Science - Artificial Intelligence, Multi-armed bandits, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract: We consider a non-stationary formulation of the stochastic multi-armed bandit where the rewards are no longer assumed to be identically distributed. For the best-arm identification task, we introduce a version of Successive Elimination based on random shuffling of the $K$ arms. We prove that under a novel and mild assumption on the mean gap $\Delta$, this simple but powerful modification achieves the same guarantees in term of sample complexity and cumulative regret than its original version, but in a much wider class of problems, as it is not anymore constrained to stationary distributions. We also show that the original {\sc Successive Elimination} fails to have controlled regret in this more general scenario, thus showing the benefit of shuffling. We then remove our mild assumption and adapt the algorithm to the best-arm identification task with switching arms. We adapt the definition of the sample complexity for that case and prove that, against an optimal policy with $N-1$ switches of the optimal arm, this new algorithm achieves an expected sample complexity of $O(\Delta^{-2}\sqrt{NK\delta^{-1} \log(K \delta^{-1})})$, where $\delta$ is the probability of failure of the algorithm, and an expected cumulative regret of $O(\Delta^{-1}{\sqrt{NTK \log (TK)}})$ after $T$ time steps.
Published: 2016

38. Low-rank Bandits with Latent Mixtures

Author: Gopalan, Aditya, Maillard, Odalric-Ambrym, Zaki, Mohammadi, Indian Institute of Science [Bangalore] (IISc Bangalore), Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, IFCAM, 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, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: FOS: Computer and information sciences, Computer Science - Learning, reinforcement learning, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], online learning, low-rank matrices, recommender systems, Multi-armed bandits, Machine Learning (cs.LG)
Abstract: We study the task of maximizing rewards from recommending items (actions) to users sequentially interacting with a recommender system. Users are modeled as latent mixtures of C many representative user classes, where each class specifies a mean reward profile across actions. Both the user features (mixture distribution over classes) and the item features (mean reward vector per class) are unknown a priori. The user identity is the only contextual information available to the learner while interacting. This induces a low-rank structure on the matrix of expected rewards r a,b from recommending item a to user b. The problem reduces to the well-known linear bandit when either user or item-side features are perfectly known. In the setting where each user, with its stochastically sampled taste profile, interacts only for a small number of sessions, we develop a bandit algorithm for the two-sided uncertainty. It combines the Robust Tensor Power Method of Anandkumar et al. (2014b) with the OFUL linear bandit algorithm of Abbasi-Yadkori et al. (2011). We provide the first rigorous regret analysis of this combination, showing that its regret after T user interactions is $\tilde O(C\sqrt{BT})$, with B the number of users. An ingredient towards this result is a novel robustness property of OFUL, of independent interest.
Published: 2016

39. Learning Iterative Processes with Recurrent Neural Networks to Correct Satellite Image Classification Maps

Author: Maggiori, Emmanuel, Charpiat, Guillaume, Tarabalka, Yuliya, Alliez, Pierre, Geometric Modeling of 3D Environments (TITANE), 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), Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Laboratoire de Recherche en Informatique (LRI), and 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
Subjects: [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]
Abstract: While initially devised for image categorization, convolutional neural networks (CNNs) are being increasingly used for the pixelwise semantic labeling of images. However, the proper nature of the most common CNN architectures makes them good at recognizing but poor at localizing objects precisely. This problem is magnified in the context of aerial and satellite image labeling, where a spatially fine object outlining is of paramount importance. Different iterative enhancement algorithms have been presented in the literature to progressively improve the coarse CNN outputs, seeking to sharpen object boundaries around real image edges. However, one must carefully design, choose and tune such algorithms. Instead, our goal is to directly learn the iterative process itself. For this, we formulate a generic iterative enhancement process inspired from partial differential equations, and observe that it can be expressed as a recurrent neural network (RNN). Consequently, we train such a network from manually labeled data for our enhancement task. In a series of experiments we show that our RNN effectively learns an iterative process that significantly improves the quality of satellite image classification maps.
Published: 2016

40. Le phénomène 'pro ana': Troubles alimentaires et réseaux sociaux

Author: Casilli, Antonio, Tubaro, Paola, Centre Edgar-Morin (CEM), Institut interdisciplinaire d'anthropologie du contemporain (IIAC), École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS), Télécom ParisTech, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Cécile Méadel, ANR-09-ALIA-0001,Ana-mia,La sociabilité 'Ana-mia': une approche des troubles alimentaires par les réseaux sociaux en ligne et hors ligne(2009), 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, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: [SHS.SOCIO]Humanities and Social Sciences/Sociology, troubles des conduites alimentaires, santé publique, "pro ana", réseaux sociaux, [SDV.SPEE]Life Sciences [q-bio]/Santé publique et épidémiologie, communautés en ligne
Abstract: Le texte ci-joint est la version pre-print de l'ouvrage, sans la postface de Cécile Méadel ni les ajouts de l'éditeur. Les numéros de page ne correspondent pas nécessairement à l'édition publiée.; International audience; Consultés et administrés par des personnes atteintes d’anorexie ou d’autres troubles alimentaires, les sites web dits « pro-ana » sont accusés d’inciter à la maigreur extrême. Depuis le début des années 2000, des vagues de polémiques amplifiées par les médias ont entouré ces blogs et forums, si bien que les pouvoirs publics ont essayé à maintes reprises de les réglementer. Mais qui sont leurs auteurs, commentateurs et lecteurs ? Que recherchent-ils vraiment ? Avec qui communiquent-ils ?Et si le « pro-ana » était moins un effet indésirable de la parole libérée sur Internet que le symptôme de transformations profondes de notre manière de vivre la santé ? Face au ralentissement des dépenses de santé publique, les patients sont aujourd’hui confrontés à l’injonction croissante d’être actifs, informés, équipés. Mais la pression sociale qui en découle peut conduire à des comportements paradoxaux et à des prises de risque.Ce livre lève le voile sur cette communauté qui a longtemps échappé au regard. Au-delà de tout propos provocateur, il interroge les usagers de ces sites internet et met au jour leurs réseaux de relations. Les résultats de cette étude défient les idées reçues et restituent une image ambivalente du web des troubles des conduites alimentaires. La valorisation de la maigreur et de l’anorexie est loin d’y faire l’unanimité. En revanche, on voit apparaître des réseaux d’entraide et d’information, dans un effort de combattre l’isolement et de pallier les manques ressentis des services de soins.
Published: 2016

41. Overcoming Calibration Problems in Pattern Labeling with Pairwise Ratings: Application to Personality Traits

Author: Isabelle Guyon, Nihar B. Shah, Marc Oliu Simon, Baiyu Chen, Víctor Ponce-López, Sergio Escalera, Lawrence Berkeley National Laboratory [Berkeley] (LBNL), Computer Vision Center (Centre de visio per computador) (CVC), Universitat Autònoma de Barcelona (UAB), Machine Learning and Optimisation (TAO), 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), Chalearn, Department of Electrical Engineering and Computer Sciences (Berkeley EECS), Gang Hua, Hervé Jégou, 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), and Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec
Subjects: Calibration (statistics), media_common.quotation_subject, Scale (descriptive set theory), 02 engineering and technology, 010501 environmental sciences, Machine learning, computer.software_genre, 01 natural sciences, Task (project management), [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], 0202 electrical engineering, electronic engineering, information engineering, Personality, 0105 earth and related environmental sciences, Mathematics, media_common, business.industry, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Function (mathematics), Regression, Variable (computer science), 020201 artificial intelligence & image processing, Pairwise comparison, Artificial intelligence, business, computer, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing
Abstract: International audience; We address the problem of calibration of workers whose task is to label patterns with continuous variables, which arises for instance in labeling images of videos of humans with continuous traits. Worker bias is particularly difficult to evaluate and correct when many workers contribute just a few labels, a situation arising typically when labeling is crowd-sourced. In the scenario of labeling short videos of people facing a camera with personality traits, we evaluate the feasibility of the pairwise ranking method to alleviate bias problems. Workers are exposed to pairs of videos at a time and must order by preference. The variable levels are reconstructed by fitting a Bradley-Terry-Luce model with maximum likelihood. This method may at first sight, seem prohibitively expensive because for N videos, p=N(N−1)/2 pairs must be potentially processed by workers rather that N videos. However, by performing extensive simulations, we determine an empirical law for the scaling of the number of pairs needed as a function of the number of videos in order to achieve a given accuracy of score reconstruction and show that the pairwise method is affordable. We apply the method to the labeling of a large scale dataset of 10,000 videos used in the ChaLearn Apparent Personality Trait challenge.
Published: 2016
Full Text: View/download PDF

42. Anti Imitation-Based Policy Learning

Author: Michèle Sebag, Marc Schoenauer, Riad Akrour, Basile Mayeur, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Paolo Frasconi, Niels Landwehr, Giuseppe Manco, Jilles Vreeken, 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, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: Value (ethics), Sequence, Computer science, business.industry, media_common.quotation_subject, Q-learning, 02 engineering and technology, 010501 environmental sciences, 01 natural sciences, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Bellman equation, Bounded function, 0202 electrical engineering, electronic engineering, information engineering, State space, 020201 artificial intelligence & image processing, Artificial intelligence, Imitation, business, Function (engineering), Mathematical economics, 0105 earth and related environmental sciences, media_common
Abstract: International audience; The Anti Imitation-based Policy Learning (AIPoL) approach, taking inspiration from the Energy-based learning framework (LeCun et al. 2006), aims at a pseudo-value function such that it induces the same order on the state space as a (nearly optimal) value function. By construction , the greedification of such a pseudo-value induces the same policy as the value function itself. The approach assumes that, thanks to prior knowledge, not-to-be-imitated demonstrations can easily be generated. For instance, applying a random policy on a good initial state (e.g., a bicycle in equilibrium) will on average lead to visit states with decreasing values (the bicycle ultimately falls down). Such a demonstration , that is, a sequence of states with decreasing values, is used along a standard learning-to-rank approach to define a pseudo-value function. If the model of the environment is known, this pseudo-value directly induces a policy by greedification. Otherwise, the bad demonstrations are exploited together with off-policy learning to learn a pseudo-Q-value function and likewise thence derive a policy by greedification. To our best knowledge the use of bad demonstrations to achieve policy learning is original. The theoretical analysis shows that the loss of optimality of the pseudo value-based policy is bounded under mild assumptions, and the empirical validation of AIPoL on the mountain car, the bicycle and the swing-up pendulum problems demonstrates the simplicity and the merits of the approach.
Published: 2016
Full Text: View/download PDF

43. Augmented Lagrangian Constraint Handling for CMA-ES---Case of a Single Linear Constraint

Author: Nikolaus Hansen, Anne Auger, Asma Atamna, Machine Learning and Optimisation (TAO), 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), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), and Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec
Subjects: Mathematical optimization, Hybrid algorithm (constraint satisfaction), 0102 computer and information sciences, 02 engineering and technology, Binary constraint, Constraint satisfaction, 01 natural sciences, Slack variable, Constraint (information theory), 010201 computation theory & mathematics, Constraint graph, Constraint logic programming, 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, [INFO]Computer Science [cs], Difference-map algorithm, Mathematics
Abstract: International audience; We consider the problem of minimizing a function f subject to a single inequality constraint g(x)
Published: 2016
Full Text: View/download PDF

44. Feature Based Algorithm Configuration: A Case Study with Differential Evolution

Author: Johann Dreo, Pierre Savéant, Marc Schoenauer, Nacim Belkhir, Machine Learning and Optimisation (TAO), 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), Thales Research and Technologies [Orsay] (TRT), THALES [France], Handl, J., Hart, E., Lewis, P.R., López-Ibáñez, M., Ochoa, G., Paechter, B., 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, and THALES
Subjects: Algorithm configuration, Empirical Study, Test bench, Mathematical optimization, Empirical Performance Model, 0102 computer and information sciences, 02 engineering and technology, 01 natural sciences, Domain (software engineering), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Black-box Continuous Optimization, Empirical research, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Robustness (computer science), Black box, Problem Fea-tures, 0202 electrical engineering, electronic engineering, information engineering, Feature based, Algorithm Configuration, Mathematics, Differential Evolution, [INFO.INFO-NA]Computer Science [cs]/Numerical Analysis [cs.NA], 010201 computation theory & mathematics, Differential evolution, 020201 artificial intelligence & image processing, [MATH.MATH-OC]Mathematics [math]/Optimization and Control [math.OC]
Abstract: International audience; Algorithm Configuration is still an intricate problem especially in the continuous black box optimization domain. This paper empirically investigates the relationship between continuous problem features (measuring different problem characteristics) and the best parameter configuration of a given stochastic algorithm over a bench of test functions — namely here, the original version of Differential Evolution over the BBOB test bench. This is achieved by learning an empirical performance model from the problem features and the algorithm parameters. This performance model can then be used to compute an empirical optimal parameter configuration from features values. The results show that reasonable performance models can indeed be learned, resulting in a better parameter configuration than a static parameter setting optimized for robustness over the test bench.
Published: 2016
Full Text: View/download PDF

45. Online Model Selection for Restricted Covariance Matrix Adaptation

Author: Nikolaus Hansen, Youhei Akimoto, Machine Learning and Optimisation (TAO), 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), J. Handl, E. Hart, P.R. Lewis, M. López-Ibáñez, G. Ochoa, B. Paechter, 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), and Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec
Subjects: Online model, Mathematical optimization, Covariance matrix, Model selection, MathematicsofComputing_NUMERICALANALYSIS, 0102 computer and information sciences, 02 engineering and technology, Function (mathematics), Covariance, 01 natural sciences, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], 010201 computation theory & mathematics, 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, [MATH.MATH-OC]Mathematics [math]/Optimization and Control [math.OC], CMA-ES, Evolution strategy, Selection (genetic algorithm), Mathematics
Abstract: International audience; We focus on a variant of covariance matrix adaptation evolution strategy (CMA-ES) with a restricted covariance matrix model, namely VkD-CMA, which is aimed at reducing the internal time complexity and the adaptation time in terms of function evaluations. We tackle the shortage of the VkD-CMA—the model of the restricted covariance matrices needs to be selected beforehand. We propose a novel mechanism to adapt the model online in the VkD-CMA. It eliminates the need for advance model selection and leads to a performance competitive with or even better than the algorithm with a nearly optimal but fixed model.
Published: 2016

46. Learning opening books in partially observable games: Using random seeds in Phantom Go

Author: Jialin Liu, Olivier Teytaud, Tristan Cazenave, Fabien Teytaud, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), University of Essex, Laboratoire d'Informatique Signal et Image de la Côte d'Opale (LISIC), Université du Littoral Côte d'Opale (ULCO), Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Teytaud, Fabien, 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, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: FOS: Computer and information sciences, Boosting (machine learning), Computer Science - Artificial Intelligence, Random seed, Monte Carlo method, 91A05, 91A10, 0102 computer and information sciences, 02 engineering and technology, 01 natural sciences, Imaging phantom, symbols.namesake, Computer Science - Computer Science and Game Theory, 0202 electrical engineering, electronic engineering, information engineering, Mathematics, [INFO.INFO-GT]Computer Science [cs]/Computer Science and Game Theory [cs.GT], Observable, Artificial Intelligence (cs.AI), 010201 computation theory & mathematics, Nash equilibrium, [INFO.INFO-GT] Computer Science [cs]/Computer Science and Game Theory [cs.GT], symbols, Probability distribution, 020201 artificial intelligence & image processing, Algorithm, Computer Science and Game Theory (cs.GT), Intuition
Abstract: Many artificial intelligences (AIs) are randomized. One can be lucky or unlucky with the random seed; we quantify this effect and show that, maybe contrarily to intuition, this is far from being negligible. Then, we apply two different existing algorithms for selecting good seeds and good probability distributions over seeds. This mainly leads to learning an opening book. We apply this to Phantom Go, which, as all phantom games, is hard for opening book learning. We improve the winning rate from 50% to 70% in 5x5 against the same AI, and from approximately 0% to 40% in 5x5, 7x7 and 9x9 against a stronger (learning) opponent., 7 pages, 15 figures. Accepted by CIG2016
Published: 2016
Full Text: View/download PDF

47. Cycle-based Cluster Variational Method for Direct and Inverse Inference

Author: Furtlehner, Cyril, Decelle, Aurélien, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, 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)-Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Inria Saclay - Ile de France, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: cluster variational method, Markov random fields, [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], generalized belief propagation, [PHYS.COND.CM-SM]Physics [physics]/Condensed Matter [cond-mat]/Statistical Mechanics [cond-mat.stat-mech]
Abstract: International audience; Large scale inference problems of practical interest can often be addressed with help of Markov random fields. This requires to solve in principle two related problems: the first one is to find offline the parameters of the MRF from empirical data (inverse problem); the second one (direct problem) is to set up the inference algorithm to make it as precise, robust and efficient as possible. In this work we address both the direct and inverse problem with mean-field methods of statistical physics, going beyond the Bethe approximation and associated belief propagation algorithm. We elaborate on the idea that loop corrections to belief propagation could be dealt with in a systematic way on pairwise Markov random fields, by using the elements of a cycle basis to define region in a generalized belief propagation setting. The region graph is specified in such a way as to avoid dual loops as much as possible, by discarding redundant Lagrange multipliers, in order to facilitate the convergence, while avoiding instabilities associated to minimal factor graph construction. We end up with a two-level algorithm, where a belief propagation algorithm is run alternatively at the level of each cycle and at the inter-region level. The inverse problem of finding the couplings of a Markov random field from empirical covariances can be addressed region wise. It turns out that this can be done efficiently in particular in the Ising context, where fixed point equations can be derived along with a one-parameter log likelihood function to minimize. Numerical experiments confirm the effectiveness of these considerations both for the direct and inverse MRF inference.; Nous examinons en détails une façon de traiter de façon systématique pour un champ Markovien aléatoire à interactions de paires,les corrections de boucles à l'algorithme de ``propagation de croyances'', consistant à choisir pour régions externes, dans un contexte de propagation de croyances généralisées,les éléments d'une base de cycles. Le graphe de régions est spécifié de telle sorte qu'on évite les boucles de rétroactions autant que faire ce peut dans le graphe des facteurs, en écartant un certain nombre decontraintes redondantes, afin de faciliter la convergence globale tout en évitantdes instabilités qui résulteraient d'une construction a minima du graphe de régions.Nous aboutissons à un algorithme de propagation de croyances a deux niveaux, celui des cycles eux-mêmes pris individuellement et celui du graphe dual où chaquenœud correspond à un cycle. Suivant cette logique, le problème inverse consistant à déterminer les paramètresde couplages du champ Markovien aléatoire à partir de covariances empiriques,factorise en une somme de problèmes inverses indépendants correspondant à chaque élément de la base de cycles. Il se trouve que ceci peut se faire efficacement, en particulier pour le problème d'Ising inverse, à l'aide d'équations de points fixes et d'optimisations de fonction de log vraisemblance à un seul paramètre. Des expériences numériques viennent conforterla pertinence de cette construction à la fois pour le problème d'inférence direct ainsique pour le problème inverse de sélection de champ Markovien aléatoire.
Published: 2016
Full Text: View/download PDF

48. Cycle-based Cluster Variational Method for Direct and Inverse Inference

Author: Aurélien Decelle, Cyril Furtlehner, Machine Learning and Optimisation (TAO), 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Laboratoire de Recherche en Informatique (LRI), and Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)
Subjects: Random field, Markov chain, Computer science, Inference, Inverse, Statistical and Nonlinear Physics, 010501 environmental sciences, Inverse problem, Condensed Matter - Disordered Systems and Neural Networks, Belief propagation, 01 natural sciences, cluster variational method, Markov random fields, [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], 0103 physical sciences, generalized belief propagation, Cycle basis, [PHYS.COND.CM-SM]Physics [physics]/Condensed Matter [cond-mat]/Statistical Mechanics [cond-mat.stat-mech], 010306 general physics, Likelihood function, Algorithm, Mathematical Physics, 0105 earth and related environmental sciences
Abstract: We elaborate on the idea that loop corrections to belief propagation could be dealt with in a systematic way on pairwise Markov random fields, by using the elements of a cycle basis to define region in a generalized belief propagation setting. The region graph is specified in such a way as to avoid dual loops as much as possible, by discarding redundant Lagrange multipliers, in order to facilitate the convergence, while avoiding instabilities associated to minimal factor graph construction. We end up with a two-level algorithm, where a belief propagation algorithm is run alternatively at the level of each cycle and at the inter-region level. The inverse problem of finding the couplings of a Markov random field from empirical covariances can be addressed region wise. It turns out that this can be done efficiently in particular in the Ising context, where fixed point equations can be derived along with a one-parameter log likelihood function to minimize. Numerical experiments confirm the effectiveness of these considerations both for the direct and inverse MRF inference., Comment: 47 pages, 16 figures
Published: 2016
Full Text: View/download PDF

49. Formalization and mathematical modelling

Author: Paola Tubaro, Machine Learning and Optimisation (TAO), 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), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Centre National de la Recherche Scientifique (CNRS), Gilbert Faccarello, Heinz Kurz, 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec, Faccarello, Gilbert, and Kurz, Heinz D.
Subjects: History of economic thought, HC, history of economics, Mathematical economics, JEL: B - History of Economic Thought, Methodology, and Heterodox Approaches/B.B2 - History of Economic Thought since 1925/B.B2.B23 - Econometrics • Quantitative and Mathematical Studies, Applied economics, HB, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, microeconomics, JEL: B - History of Economic Thought, Methodology, and Heterodox Approaches/B.B4 - Economic Methodology/B.B4.B41 - Economic Methodology, Economics, Social science, Positive economics, QA, JEL: B - History of Economic Thought, Methodology, and Heterodox Approaches/B.B2 - History of Economic Thought since 1925/B.B2.B21 - Microeconomics, ComputingMilieux_MISCELLANEOUS, JEL: B - History of Economic Thought, Methodology, and Heterodox Approaches/B.B1 - History of Economic Thought through 1925/B.B1.B16 - Quantitative and Mathematical
Abstract: Claims that economics is close in spirit and principle to mathematics have been reiterated many times; a well-known one is William S. Jevons’s “our science must be mathematical, simply because it deals with quantities” (1871 [1888]: 1.5, original emphasis). The use of mathematics is now widely recognized as an essential ingredient of both research and teaching in economics, and attempts to build mathematical accounts of economic phenomena are as old as the discipline itself (Theocharis 1961 [1983]). The history of mathematical modelling in economics has not been a linear one. The profession privileged the verbal form for long, and even when the formal approach gradually gained ground, it often had to defend itself against hefty criticisms. Today’s widespread consensus around formalization does not prevent doubts from occasionally resurfacing.
Published: 2016
Full Text: View/download PDF

50. Analysis of Different Types of Regret in Continuous Noisy Optimization

Author: Sandra Astete-Morales, Marie-Liesse Cauwet, Olivier Teytaud, Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Machine Learning and Optimisation (TAO), 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), T. Friedrich and F. Neumann, 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)-Université Paris-Sud - Paris 11 (UP11)-Laboratoire de Recherche en Informatique (LRI), and Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec
Subjects: Computer Science::Machine Learning, Sequence, Mathematical optimization, Optimization algorithm, Regret, 0102 computer and information sciences, 02 engineering and technology, 01 natural sciences, Measure (mathematics), Noisy optimization, Term (time), Statistics::Machine Learning, Rate of convergence, 010201 computation theory & mathematics, Simple (abstract algebra), Optimization and Control (math.OC), 0202 electrical engineering, electronic engineering, information engineering, FOS: Mathematics, 020201 artificial intelligence & image processing, [MATH.MATH-OC]Mathematics [math]/Optimization and Control [math.OC], Sphere function, Mathematics - Optimization and Control, performance measure, regret analysis, Mathematics
Abstract: The performance measure of an algorithm is a crucial part of its analysis. The performance can be determined by the study on the convergence rate of the algorithm in question. It is necessary to study some (hopefully convergent) sequence that will measure how "good" is the approximated optimum compared to the real optimum. The concept of Regret is widely used in the bandit literature for assessing the performance of an algorithm. The same concept is also used in the framework of optimization algorithms, sometimes under other names or without a specific name. And the numerical evaluation of convergence rate of noisy algorithms often involves approximations of regrets. We discuss here two types of approximations of Simple Regret used in practice for the evaluation of algorithms for noisy optimization. We use specific algorithms of different nature and the noisy sphere function to show the following results. The approximation of Simple Regret, termed here Approximate Simple Regret, used in some optimization testbeds, fails to estimate the Simple Regret convergence rate. We also discuss a recent new approximation of Simple Regret, that we term Robust Simple Regret, and show its advantages and disadvantages., Genetic and Evolutionary Computation Conference 2016, Jul 2016, Denver, United States. 2016
Published: 2016

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

Journal

Database

Publisher

597 results on '"Machine Learning and Optimisation (TAO)"'

Search Results

Catalog

Select search scope, currently: Articles

Catalog

books, media & more in Jio Institute collections

Articles

journal articles & other e-resources