Improving Training Result of Partially Observable Markov Decision Process by Filtering Beliefs

TL;DR

This paper introduces a belief filtering method for POMDPs that reduces training time and improves training quality by removing similar beliefs, demonstrating superior performance over existing point-based methods.

Contribution

The paper proposes a novel belief filtering approach that enhances POMDP training efficiency and effectiveness by eliminating redundant beliefs based on similarity.

Findings

Outperforms point-based approximate POMDPs in training quality

Reduces training time through belief filtering

Improves control policy performance in POMDPs

Abstract

In this study I proposed a filtering beliefs method for improving performance of Partially Observable Markov Decision Processes(POMDPs), which is a method wildly used in autonomous robot and many other domains concerning control policy. My method search and compare every similar belief pair. Because a similar belief have insignificant influence on control policy, the belief is filtered out for reducing training time. The empirical results show that the proposed method outperforms the point-based approximate POMDPs in terms of the quality of training results as well as the efficiency of the method.