Skip to Main content Skip to Navigation
Theses

Apprentissage séquentiel budgétisé pour la classification extrême et la découverte de hiérarchie en apprentissage par renforcement

Abstract : 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.
Complete list of metadatas

Cited literature [100 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-02954134
Contributor : Abes Star :  Contact
Submitted on : Wednesday, September 30, 2020 - 6:16:57 PM
Last modification on : Wednesday, November 11, 2020 - 3:21:18 AM

File

LEON_Aurelia_2019.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-02954134, version 1

Citation

Aurélia Léon. Apprentissage séquentiel budgétisé pour la classification extrême et la découverte de hiérarchie en apprentissage par renforcement. Intelligence artificielle [cs.AI]. Sorbonne Université, 2019. Français. ⟨NNT : 2019SORUS226⟩. ⟨tel-02954134⟩

Share