Kernel-Based Function Approximation for Average Reward Reinforcement Learning: An Optimist No-Regret Algorithm

TL;DR

This work considers kernel-based function approximation for RL in the infinite horizon average reward setting and proposes an optimistic algorithm, similar to acquisition function based algorithms in the special case of bandits, under kernel-based modelling assumptions.

Abstract

Reinforcement learning utilizing kernel ridge regression to predict the expected value function represents a powerful method with great representational capacity. This setting is a highly versatile framework amenable to analytical results. We consider kernel-based function approximation for RL in the infinite horizon average reward setting, also referred to as the undiscounted setting. We propose an optimistic algorithm, similar to acquisition function based algorithms in the special case of bandits. We establish novel no-regret performance guarantees for our algorithm, under kernel-based modelling assumptions. Additionally, we derive a novel confidence interval for the kernel-based prediction of the expected value function, applicable across various RL problems.

Theorems & Definitions (13)

• Definition 1
• Theorem 1: Confidence Bound
• Remark 2
• Theorem 3
• Remark 4
• Lemma 1: Proposition $1$ in vakili2021optimal
• Lemma 2
• Lemma 3
• Lemma 4
• proof : Proof of Lemma \ref{['lem:ourellipt']}