Code associated with our paper Estimating Risk and Uncertainty in Deep Reinforcement Learning
Requirements can be found in the requirements.txt file.
The project contains one folder per experiment in the paper, except for Atari experiments that are all within the Atari folder. Most folders contain a jupyter notebook that can be used to reproduce the corresponding experiment, except for Bandits and Atari where the associated computation cost is quite large. Each folder contains its own README.
A previous version of this code, associated with an old version of our paper, can be found in the Archive folder.
If you find this work or the associated paper useful, it can be cited as below.
@article{clements2019risk,
title={Estimating Risk and Uncertainty in Deep Reinforcement Learning},
author={Clements, William R and Robaglia, Benoit-Marie and Van Delft, Bastien and Slaoui, Reda Bahi and Toth, Sebastien},
journal={arXiv preprint arXiv:1905.09638},
year={2019}
}