Atari 2600 games

The dataset used in this paper is a collection of state-action pairs generated by a pre-trained RL agent, used to train a self-supervised interpretable network (SSINet) to produce attention masks for explaining the agent's decisions.

Data and Resources

Cite this as

Wenjie Shi, Gao Huang, Shiji Song, Zhuoyuan Wang, Tingyu Lin, Cheng Wu (2024). Dataset: Atari 2600 games. https://doi.org/10.57702/z2n4g3e0

DOI retrieved: December 16, 2024

Additional Info

Field Value
Created December 16, 2024
Last update December 16, 2024
Defined In https://doi.org/10.48550/arXiv.2310.17173
Citation
  • https://doi.org/10.48550/arXiv.2003.07069
Author Wenjie Shi
More Authors
Gao Huang
Shiji Song
Zhuoyuan Wang
Tingyu Lin
Cheng Wu
Homepage https://github.com/shiwj16/SSINet