Action Elimination and Stopping Conditions for Reinforcement Learning

Proceedings of the Twentieth International Conference on Machine Learning

Volume

Proceedings of the Twentieth International Conference on Machine Learning

Contents

We consider incorporating action elimination procedures in reinforcement learning algorithms. We suggest a framework that is based on learning an upper and a lower estimates of the value function or the Q-function and eliminating actions that are not optimal. We provide a model-based and a model-free variants of the elimination method. We further derive stopping conditions that guarantee that the learned policy is approximately optimal with high probability. Simulations demonstrate a considerable speedup and added robustness.