LASSERRE Marvin
Team : DECISION
https://www.researchgate.net/profile/Marvin-Lasserre
Supervision : Christophe GONZALES
Co-supervision : WUILLEMIN Pierre-Henri, LEBRUN Régis
Learning non-parametric Copula Bayesian Networks
Modeling multivariate continuous distributions is a task of central interest in statistics and machine learning
with many applications in science and engineering.
However, high-dimensional distributions are difficult to handle and can lead to intractable computations.
The Copula Bayesian Networks (CBNs) take advantage of both Bayesian networks (BNs) and copula
theory to compactly represent such multivariate distributions.
Bayesian networks rely on conditional independences in order to reduce the complexity of the problem, while
copula functions allow modeling the dependence relation between random variables.
The goal of this thesis is to give a common framework for both domains and to propose new learning algorithms for copula
Bayesian networks. To do so, we use the fact that CBNs have the same graphical language as
BNs which allow us to adapt their learning methods to this model.
Moreover, using the empirical Bernstein copula both to design conditional independence tests and to estimate copulas from data,
we avoid making parametric assumptions, which gives greater generality to our methods.
Defence : 03/11/2022
Jury members :
Sébastien Destercke, chargé de recherche, CNRS, Université de Technologie de Compiègne [rapporteur]
Simon de Givry, chargé de recherche, CNRS, INRAE MIAT [rapporteur]
Gregory Nuel, Directeur de recherche, CNRS, Sorbonne Université
Patrice Perny, Professeur, Sorbonne Université
Clémentine Prieur, Professeur, Université Grenoble Alpes
Christophe Gonzales, Professeur, Université Aix Marseille
Pierre-Henri Wuillemin, Maitre de conférence, Sorbonne Université
Régis Lebrun, Senior scientist, Airbus Central Research & Technology
2018-2024 Publications
-
2024
- M. Lasserre, R. Lebrun, P.‑H. Wuillemin : “Quadrature Rules in General Continuous Bayesian Networks : Discrete Inference without Discretization”, (2024)
-
2022
- M. Lasserre : “Apprentissages dans les Réseaux Bayésiens à Base de Copules Non-Paramétriques”, thesis, phd defence 03/11/2022, supervision Gonzales, Christophe, co-supervision : Wuillemin, Pierre-Henri, Lebrun, Régis (2022)
-
2021
- M. Lasserre, R. Lebrun, P.‑H. Wuillemin : “Apprentissage de modèles continus à grandes dimensions en utilisant l’information mutuelle et les réseaux bayésiens de copules”, 10es Journées Francophones sur les Réseaux Bayésiens et les Modèles Graphiques Probabilistes, Porquerolles, France (2021)
- M. Lasserre, R. Lebrun, P.‑H. Wuillemin : “Constraint-based learning for non-parametric continuous bayesian networks”, Annals of Mathematics and Artificial Intelligence, vol. 89, pp. 1035-1052, (Springer Verlag) (2021)
- M. Lasserre, R. Lebrun, P.‑H. Wuillemin : “Learning Continuous High-Dimensional Models using Mutual Information and Copula Bayesian Networks”, Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35 (13), AAAI-21 Technical Tracks 13, Vancouver, Canada, pp. 12139-12146 (2021)
-
2020
- M. Lasserre, R. Lebrun, P.‑H. Wuillemin : “Constraint-Based Learning for Non-Parametric Continuous Bayesian Networks”, FLAIRS 33 - 33rd Florida Artificial Intelligence Research Society Conference, Miami, United States, pp. 581-586, (AAAI) (2020)
-
2018
- A. Goareguer, A. Le Denn, F. Renaude, Ch. Maragna, P.‑H. Wuillemin, M. Lasserre, Ch. Gonzales, M. Clausse : “Projet SunSTONE : réseaux de chaleur solaires intelligents avec stockage intersaisonnier”, Journées Nationales sur l'Energie Solaire - JNES 2018, Lyon, France (2018)