Optimal Dynamic Regret by Transformers for Non-Stationary Reinforcement Learning

Baiyuan Chen, Shinji Ito, Masaaki Imaizumi

Keywords: transformer, in-context learning, reinforcement learning, bandit problem, non-stationary

TL;DR: We prove that transformers can achieve nearly optimal dynamic regret bounds in non-stationary environments.

Abstract: Transformers have demonstrated exceptional performance across a wide range of domains. While their ability to perform reinforcement learning in-context has been established both theoretically and empirically, their behavior in non-stationary environments remains less understood. In this study, we address this gap by showing that transformers can achieve nearly optimal dynamic regret bounds in non-stationary settings. We prove that transformers are capable of approximating strategies used to handle non-stationary environment, and can learn the approximator in the in-context learning setup. Our experiments further show that transformers can match or even outperform existing expert algorithms in such environments.

Primary Area: Reinforcement learning (e.g., decision and control, planning, hierarchical RL, robotics)

Submission Number: 14755