Supervision : Ludovic DENOYER
Budgeted sequential learning for extreme classification and for the discovery of hierarchy in reinforcement learning
This thesis deals with the notion of budget to study problems of complexity (it can be computational complexity, a complex task for an agent, or complexity due to a small amount of data). Indeed, the main goal of current techniques in machine learning is usually to obtain the best accuracy, without worrying about the cost of the task. The concept of budget makes it possible to take into account this parameter while maintaining good performances.
We first focus on classification problems with a large number of classes: the complexity in those algorithms can be reduced thanks to the use of decision trees (here learned through budgeted reinforcement learning techniques) or the association of each class with a (binary) code. We then deal with reinforcement learning problems and the discovery of a hierarchy that breaks down a (complex) task into simpler tasks to facilitate learning and generalization. Here, this discovery is done by reducing the cognitive effort of the agent (considered in this work as equivalent to the use of an additional observation). Finally, we address problems of understanding and generating instructions in natural language, where data are available in small quantities: we test for this purpose the simultaneous use of an agent that understands and of an agent that generates the instructions.
Defence : 05/10/2019 - 13h30 - Campus Jussieu 25-26/105
Jury members :
M. Jeremie Mary, Criteo [rapporteur]
Mme. Cecile Capponi, Aix-Marseille Université - LIS [rapporteuse]
Mme. Aurélie Beynier, Sorbonne Université - LIP6
M. Stéphane Doncieux, Sorbonne Université - ISIR
M. Yves Grandvalet, Université de Technologie de Compiègne, Heudiasyc
M. Ludovic Denoyer, Facebook
- A. Léon : “Apprentissage séquentiel budgétisé pour la classification extrême et la découverte de hiérarchie en apprentissage par renforcement”, thesis, defence 05/10/2019, supervision Denoyer, Ludovic (2019)
- A. Léon, L. Denoyer : “Policy-gradient methods for Decision Trees”, ESANN - European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Unknown, (2016)
- A. Léon, L. Denoyer : “Reinforced Decision Trees”, European Workshop on Reinforcement Learning - EWRL, Unknown, (2015)