Model Checking Finite-Horizon Markov Chains with Probabilistic Inference

TL;DR

This paper explores a novel approach to verifying finite-horizon Markov chains using probabilistic inference, introducing a new tool that improves scalability over traditional model checkers.

Contribution

It formally compares traditional and inference-based methods, and develops Rubicon, a tool integrating probabilistic inference into model checking.

Findings

Rubicon demonstrates better scalability on benchmarks.

Probabilistic inference offers a promising alternative for model checking.

The approach effectively integrates symbolic path representation with existing tools.

Abstract

We revisit the symbolic verification of Markov chains with respect to finite horizon reachability properties. The prevalent approach iteratively computes step-bounded state reachability probabilities. By contrast, recent advances in probabilistic inference suggest symbolically representing all horizon-length paths through the Markov chain. We ask whether this perspective advances the state-of-the-art in probabilistic model checking. First, we formally describe both approaches in order to highlight their key differences. Then, using these insights we develop Rubicon, a tool that transpiles Prism models to the probabilistic inference tool Dice. Finally, we demonstrate better scalability compared to probabilistic model checkers on selected benchmarks. All together, our results suggest that probabilistic inference is a valuable addition to the probabilistic model checking portfolio -- with Rubicon as a first step towards integrating both perspectives.