Minimax Optimal and Computationally Efficient Algorithms for Distributionally Robust Offline Reinforcement Learning

[NeurIPS 2024]

Zhishuai Liu · Pan Xu

Duke University

Official implementation of the paper "Minimax Optimal and Computationally Efficient Algorithms for Distributionally Robust Offline Reinforcement Learning", which is published in the Proceedings of the Thirty-Eighth Annual Conference on Neural Information Processing Systems (NeurIPS).

Installation instructions

Dependencies

• python == 3.7
• scipy == 1.7.3
• matplotlib == 2.2.3
• numpy == 1.21.6

Citation

@inproceedings{liu2024minimax,
 title = {Minimax Optimal and Computationally Efficient Algorithms for Distributionally Robust Offline Reinforcement Learning},
 author = {Liu, Zhishuai and Xu, Pan},
 booktitle = {Advances in Neural Information Processing Systems},
 pages = {86602--86654},
 volume = {37},
 year = {2024}
}