Provable Policy Gradient for Robust Average-Reward MDPs Beyond Rectangularity.

Qiuhao Wang, Yuqi Zha, Chin Pang Ho, Marek Petrik

Browse the full ICML paper archive.