Deep reinforcement learning

Abstract

This seminar presents the main approaches to reinforcement learning, focusing on effective algorithms, and comparing the advantages of different approaches (with world model, without model, etc.). Yann Olivier discusses the state of mathematical understanding of these algorithms beyond toy problems, as well as both theoretical and practical limitations of current approaches.

Abstract

Documents and media

Speaker(s)

Yann Ollivier

Events

Introduction to deep neural networks

Introducing 7 data challenges 2019 (1)

Applications of deep neural networks

Introducing 7 data challenges 2019 (2)

Neural network approximations and regularity

Presentation of the 2018 challenge winners

The origins : cybernetics and the perceptron

Weakly supervised learning for visual recognition

Universal single-layer hidden network approximation

Natural language

Approximation error with a hidden layer and regularity

Automatic video analysis

Maximum likelihood network optimization

Deep reinforcement learning

Gradient descent and gradient backpropagation

Convergence of stochastic gradient descent

See also