On Solving a Stochastic Shortest-Path Markov Decision Process as Probabilistic Inference

TL;DR

This paper proposes a novel approach to solving stochastic shortest-path Markov decision processes by framing them as probabilistic inference, addressing both online and offline planning under indefinite horizons.

Contribution

It introduces a probabilistic inference framework for SSP MDPs, extending planning methods beyond finite horizons and comparing different solution approaches.

Findings

Framework unifies finite and infinite horizon MDPs

Enables online and offline planning under uncertainty

Highlights differences between dynamic programming and active inference methods

Abstract

Previous work on planning as active inference addresses finite horizon problems and solutions valid for online planning. We propose solving the general Stochastic Shortest-Path Markov Decision Process (SSP MDP) as probabilistic inference. Furthermore, we discuss online and offline methods for planning under uncertainty. In an SSP MDP, the horizon is indefinite and unknown a priori. SSP MDPs generalize finite and infinite horizon MDPs and are widely used in the artificial intelligence community. Additionally, we highlight some of the differences between solving an MDP using dynamic programming approaches widely used in the artificial intelligence community and approaches used in the active inference community.