Exploiting symmetries in POMDPs for point-based algorithms

Abstract

We extend the model minimization technique for partially observable Markov decision processes (POMDPs) to handle symmetries in the joint space of states, actions, and observations. The POMDP symmetry we de-ne in this paper cannot be handled by the model minimization techniques previously published in the literature. We formulate the problem of nding the symmetries as a graph automorphism (GA) problem, and although not yet known to be tractable, we experimentally show that the sparseness of the graph representing the POMDP allows us to quickly nd symmetries. We show how the symmetries in POMDPs can be exploited for speeding up point-based algorithms. We experimentally demonstrate the effectiveness of our approach.