Improving Training Result of Partially Observable Markov Decision Process by Filtering Beliefs
This work addresses the problem of improving the training efficiency and quality of POMDPs for applications like autonomous robots, representing an incremental improvement over existing methods.
This study proposes a filtering beliefs method to improve the training results of Partially Observable Markov Decision Processes (POMDPs). By identifying and filtering out similar belief pairs that have insignificant influence on control policy, the method reduces training time and empirically outperforms point-based approximate POMDPs in both training quality and efficiency.
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.