# Corentin Tallec

I'm a third year PhD student in machine learning, supervised by
Yann Ollivier, in the TAU team, at the Laboratoire de Recherche en
Informatique (LRI) of Saclay.

## Interests

- Recurrent neural networks
- Reinforcement learning
- Theory of deep learning
- Generative models

## Publications

- C. Tallec, L. Blier, Y. Ollivier.
**Making Deep Q-learning methods robust to time discretization**, Accepted at ICML 2019 [arxiv]
- C. Tallec, L. Blier, D. Kalainathan.
**[RE] Recurrent World Models Facilitate Policy Evolution**, reproduction paper [ReScience submission]
- C. Tallec, Y. Ollivier.
**Can recurrent neural networks warp time**, Poster ICLR 2018 [openreview]
- C. Tallec, Y. Ollivier.
**Unbiased online recurrent optimization**, Poster ICLR 2018 [arxiv]
- T. Lucas, C. Tallec, J. Verbeek, Y. Ollivier.
**Mixed batches and symmetric discriminators for GAN training**, long oral ICML 2018 [arxiv]
- C. Tallec, Y. Ollivier.
**Unbiasing truncated backpropagation through time** [arxiv]
- Y. Ollivier, C. Tallec, G. Charpiat.
**Training recurrent networks online without backtracking** [arxiv]

## Projects

**Pyvarinf**, a python package to facilitate the use of Bayesian Deep Learning in Pytorch [github]
**Reproducing World Models**, a pytorch reimplementation of the paper World Models (Ha et al. 2018) (see the related ReScience submission) [github]