Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Yann Ollivier

Unbiased Methods for Multi-Goal Reinforcement Learning


Jun 16, 2021
LĂ©onard Blier, Yann Ollivier

* 9 pages 

  Access Paper or Ask Questions

Learning One Representation to Optimize All Rewards


Mar 14, 2021
Ahmed Touati, Yann Ollivier


  Access Paper or Ask Questions

Learning Successor States and Goal-Dependent Values: A Mathematical Viewpoint


Jan 18, 2021
LĂ©onard Blier, Corentin Tallec, Yann Ollivier


  Access Paper or Ask Questions

Convergence of Online Adaptive and Recurrent Optimization Algorithms


May 12, 2020
Pierre-Yves Massé, Yann Ollivier


  Access Paper or Ask Questions

Interpreting a Penalty as the Influence of a Bayesian Prior


Feb 01, 2020
Pierre Wolinski, Guillaume Charpiat, Yann Ollivier

* 24 pages, including 2 pages of references and 10 pages of appendix 

  Access Paper or Ask Questions

White-box vs Black-box: Bayes Optimal Strategies for Membership Inference


Aug 29, 2019
Alexandre Sablayrolles, Matthijs Douze, Yann Ollivier, Cordelia Schmid, Hervé Jégou


  Access Paper or Ask Questions

Separating value functions across time-scales


Feb 08, 2019
Joshua Romoff, Peter Henderson, Ahmed Touati, Yann Ollivier, Emma Brunskill, Joelle Pineau


  Access Paper or Ask Questions

Making Deep Q-learning methods robust to time discretization


Jan 29, 2019
Corentin Tallec, LĂ©onard Blier, Yann Ollivier


  Access Paper or Ask Questions

The Extended Kalman Filter is a Natural Gradient Descent in Trajectory Space


Jan 03, 2019
Yann Ollivier


  Access Paper or Ask Questions

The Description Length of Deep Learning Models


Nov 01, 2018
LĂ©onard Blier, Yann Ollivier

* NIPS 2018 

  Access Paper or Ask Questions

Adversarial Vulnerability of Neural Networks Increases With Input Dimension


Oct 08, 2018
Carl-Johann Simon-Gabriel, Yann Ollivier, Léon Bottou, Bernhard Schölkopf, David Lopez-Paz

* 10 pages main text and references, 8 pages appendix, 7 figures 

  Access Paper or Ask Questions

Learning with Random Learning Rates


Oct 03, 2018
LĂ©onard Blier, Pierre Wolinski, Yann Ollivier

* 20 pages, 8 figures, code available on GitHub 

  Access Paper or Ask Questions

Online Natural Gradient as a Kalman Filter


Aug 27, 2018
Yann Ollivier

* 3rd version: expanded intro 

  Access Paper or Ask Questions

Mixed batches and symmetric discriminators for GAN training


Jun 19, 2018
Thomas Lucas, Corentin Tallec, Jakob Verbeek, Yann Ollivier

* Accepted at ICML 2018 (long oral) 

  Access Paper or Ask Questions

Approximate Temporal Difference Learning is a Gradient Descent for Reversible Policies


May 02, 2018
Yann Ollivier


  Access Paper or Ask Questions

Can recurrent neural networks warp time?


Mar 23, 2018
Corentin Tallec, Yann Ollivier


  Access Paper or Ask Questions

True Asymptotic Natural Gradient Optimization


Dec 22, 2017
Yann Ollivier


  Access Paper or Ask Questions

Natural Langevin Dynamics for Neural Networks


Dec 04, 2017
Gaétan Marceau-Caron, Yann Ollivier


  Access Paper or Ask Questions

Unbiasing Truncated Backpropagation Through Time


May 23, 2017
Corentin Tallec, Yann Ollivier


  Access Paper or Ask Questions

Unbiased Online Recurrent Optimization


May 23, 2017
Corentin Tallec, Yann Ollivier

* 11 pages, 5 figures 

  Access Paper or Ask Questions

Practical Riemannian Neural Networks


Feb 25, 2016
Gaétan Marceau-Caron, Yann Ollivier


  Access Paper or Ask Questions

Training recurrent networks online without backtracking


Nov 20, 2015
Yann Ollivier, Corentin Tallec, Guillaume Charpiat


  Access Paper or Ask Questions

Speed learning on the fly


Nov 08, 2015
Pierre-Yves Massé, Yann Ollivier

* preprint 

  Access Paper or Ask Questions

Riemannian metrics for neural networks II: recurrent networks and learning symbolic data sequences


Feb 03, 2015
Yann Ollivier

* 4th version: some changes in notation, more experiments 

  Access Paper or Ask Questions

Riemannian metrics for neural networks I: feedforward networks


Feb 03, 2015
Yann Ollivier

* (5th version, minor changes) 

  Access Paper or Ask Questions

Auto-encoders: reconstruction versus compression


Jan 23, 2015
Yann Ollivier


  Access Paper or Ask Questions

Objective Improvement in Information-Geometric Optimization


Mar 07, 2013
Youhei Akimoto, Yann Ollivier

* Foundations of Genetic Algorithms XII (2013) 

  Access Paper or Ask Questions

Layer-wise learning of deep generative models


Feb 16, 2013
Ludovic Arnold, Yann Ollivier


  Access Paper or Ask Questions