Online Markov Decision Processes with Non-oblivious Strategic Adversary

TL;DR

This paper introduces algorithms for Online Markov Decision Processes against strategic adversaries, achieving regret bounds and convergence results, especially in large action spaces and game-theoretic settings.

Contribution

It extends existing algorithms to handle non-oblivious adversaries, proposes MDP-OOE leveraging Double Oracle, and provides the first last-iteration convergence result in OMDPs.

Findings

MDP-Expert achieves regret bounds with oblivious adversaries.

MDP-OOE effectively handles large action spaces using game-theoretic ideas.

First last-round convergence result to NE in OMDPs.

Abstract

We study a novel setting in Online Markov Decision Processes (OMDPs) where the loss function is chosen by a non-oblivious strategic adversary who follows a no-external regret algorithm. In this setting, we first demonstrate that MDP-Expert, an existing algorithm that works well with oblivious adversaries can still apply and achieve a policy regret bound of $\mathcal{O}(\sqrt{T \log(L)}+\tau^2\sqrt{ T \log(|A|)})$ where $L$ is the size of adversary's pure strategy set and $|A|$ denotes the size of agent's action space. Considering real-world games where the support size of a NE is small, we further propose a new algorithm: MDP-Online Oracle Expert (MDP-OOE), that achieves a policy regret bound of $\mathcal{O}(\sqrt{T\log(L)}+\tau^2\sqrt{ T k \log(k)})$ where $k$ depends only on the support size of the NE. MDP-OOE leverages the key benefit of Double Oracle in game theory and thus can solve games with prohibitively large action space. Finally, to better understand the learning dynamics of no-regret methods, under the same setting of no-external regret adversary in OMDPs, we introduce an algorithm that achieves last-round convergence result to a NE. To our best knowledge, this is first work leading to the last iteration result in OMDPs.