Improved Training Mechanism for Reinforcement Learning via Online Model Selection

TL;DR

This paper introduces an online model selection approach for reinforcement learning that adaptively chooses the best agent configuration, improving training efficiency and stability through theoretical insights and empirical validation.

Contribution

It presents a novel online model selection framework for reinforcement learning that enhances efficiency, adaptability, and stability across various agent configurations.

Findings

Improved resource allocation in RL training.

Enhanced adaptation to non-stationary environments.

Increased training stability across different seeds.

Abstract

We study the problem of online model selection in reinforcement learning, where the selector has access to a class of reinforcement learning agents and learns to adaptively select the agent with the right configuration. Our goal is to establish the improved efficiency and performance gains achieved by integrating online model selection methods into reinforcement learning training procedures. We examine the theoretical characterizations that are effective for identifying the right configuration in practice, and address three practical criteria from a theoretical perspective: 1) Efficient resource allocation, 2) Adaptation under non-stationary dynamics, and 3) Training stability across different seeds. Our theoretical results are accompanied by empirical evidence from various model selection tasks in reinforcement learning, including neural architecture selection, step-size selection, and self model selection.