Distribution over Beliefs for Memory Bounded Dec-POMDP Planning

TL;DR

This paper introduces a novel point-based approximate planning method for Dec-POMDPs that improves solution quality by heuristically estimating belief priors and optimizing policy tree selection to reduce pruning errors.

Contribution

It presents a new heuristic-based approach for Dec-POMDP planning that outperforms existing methods in solution quality.

Findings

Outperforms state-of-the-art Dec-POMDP planning methods

Uses heuristic estimation of belief priors for policy selection

Formulates policy tree selection as a combinatorial optimization problem

Abstract

We propose a new point-based method for approximate planning in Dec-POMDP which outperforms the state-of-the-art approaches in terms of solution quality. It uses a heuristic estimation of the prior probability of beliefs to choose a bounded number of policy trees: this choice is formulated as a combinatorial optimisation problem minimising the error induced by pruning.