Model-based Offline Policy Optimization with Adversarial Network (MOAN)

Offline RL framework called MOAN, which introduces a two-player game to improve the generalization capability of the transition model and mitigate the negative effects of potentially problematic rollouts during offline reinforcement learning.

Data and Resources

Cite this as

Junming Yang, Xingguo Chen, Shengyuan Wang, Bolei Zhang (2025). Dataset: Model-based Offline Policy Optimization with Adversarial Network (MOAN). https://doi.org/10.57702/dv0g8aos

DOI retrieved: January 2, 2025

Additional Info

Field Value
Created January 2, 2025
Last update January 2, 2025
Defined In https://doi.org/10.48550/arXiv.2309.02157
Author Junming Yang
More Authors
Xingguo Chen
Shengyuan Wang
Bolei Zhang
Homepage https://github.com/junming-yang/MOAN