Kernel-Based Function Approximation for Average Reward Reinforcement Learning: An Optimist No-Regret Algorithm
Keywords: Reinforcement learning, infinite horizon average reward setting, no-regret algorithm, kernel-based model
TL;DR: We propose an optimisitc kernel-based RL algorithm for the infinite horizon average reward setting and prove no-regret performance guarantees.
Abstract: Reinforcement Learning (RL) 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.
Primary Area: Reinforcement learning
Submission Number: 7056
Loading