Timezone: »

Improved Corruption Robust Algorithms for Episodic Reinforcement Learning
Yifang Chen · Simon Du · Kevin Jamieson

Wed Jul 21 05:35 PM -- 05:40 PM (PDT) @
We study episodic reinforcement learning under unknown adversarial corruptions in both the rewards and the transition probabilities of the underlying system. We propose new algorithms which, compared to the existing results in \cite{lykouris2020corruption}, achieve strictly better regret bounds in terms of total corruptions for the tabular setting. To be specific, firstly, our regret bounds depend on more precise numerical values of total rewards corruptions and transition corruptions, instead of only on the total number of corrupted episodes. Secondly, our regret bounds are the first of their kind in the reinforcement learning setting to have the number of corruptions show up additively with respect to $\min\{ \sqrt{T},\text{PolicyGapComplexity} \}$ rather than multiplicatively. Our results follow from a general algorithmic framework that combines corruption-robust policy elimination meta-algorithms, and plug-in reward-free exploration sub-algorithms. Replacing the meta-algorithm or sub-algorithm may extend the framework to address other corrupted settings with potentially more structure.

Author Information

Yifang Chen (University of Washington)
Simon Du (University of Washington)
Kevin Jamieson (University of Washington)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors