PAC-Bayesian Policy Evaluation for Reinforcement Learning

TL;DR

This paper introduces a PAC-Bayesian bound for batch reinforcement learning with function approximation, enabling robust policy evaluation that effectively leverages informative priors and ignores misleading ones.

Contribution

It presents the first PAC-Bayesian bound for batch RL with function approximation and demonstrates its use in model selection within transfer learning.

Findings

PAC-Bayesian policy evaluation effectively uses informative priors

The method ignores misleading priors, improving robustness

Empirical results validate the approach's effectiveness

Abstract

Bayesian priors offer a compact yet general means of incorporating domain knowledge into many learning tasks. The correctness of the Bayesian analysis and inference, however, largely depends on accuracy and correctness of these priors. PAC-Bayesian methods overcome this problem by providing bounds that hold regardless of the correctness of the prior distribution. This paper introduces the first PAC-Bayesian bound for the batch reinforcement learning problem with function approximation. We show how this bound can be used to perform model-selection in a transfer learning scenario. Our empirical results confirm that PAC-Bayesian policy evaluation is able to leverage prior distributions when they are informative and, unlike standard Bayesian RL approaches, ignore them when they are misleading.