Utilization of domain knowledge to improve POMDP belief estimation

TL;DR

This paper introduces a novel method that incorporates domain knowledge into POMDP belief updates using Jeffrey's rule, enhancing policy learning efficiency and performance in decision-making under uncertainty.

Contribution

The paper presents a new approach for integrating domain knowledge into POMDP belief estimation via Jeffrey's rule, reducing data needs and improving RL policy performance.

Findings

Domain knowledge integration improves belief estimation accuracy.

Reduces data requirements for effective POMDP policy learning.

Enhances RL policy performance in uncertain environments.

Abstract

The partially observable Markov decision process (POMDP) framework is a common approach for decision making under uncertainty. Recently, multiple studies have shown that by integrating relevant domain knowledge into POMDP belief estimation, we can improve the learned policy's performance. In this study, we propose a novel method for integrating the domain knowledge into probabilistic belief update in POMDP framework using Jeffrey's rule and normalization. We show that the domain knowledge can be utilized to reduce the data requirement and improve performance for POMDP policy learning with RL.