Hypertree Decompositions Revisited for PGMs

TL;DR

This paper introduces JoinInfer, an exact inference engine for PGMs based on worst-case optimal join algorithms, demonstrating significant empirical speedups over existing methods and proposing a data-driven heuristic for adaptive performance.

Contribution

The paper presents the first empirical evaluation of worst-case optimal join algorithms for PGM inference, outperforming state-of-the-art engines and introducing a heuristic for automatic parameter tuning.

Findings

JoinInfer outperforms ACE, IJGP, libDAI by up to 630x on benchmarks.

Empirical analysis refines theoretical notions of data properties affecting inference.

A data-driven heuristic enables automatic adaptation of inference parameters.

Abstract

We revisit the classical problem of exact inference on probabilistic graphical models (PGMs). Our algorithm is based on recent \emph{worst-case optimal database join} algorithms, which can be asymptotically faster than traditional data processing methods. We present the first empirical evaluation of these algorithms via JoinInfer -- a new exact inference engine. We empirically explore the properties of the data for which our engine can be expected to outperform traditional inference engines, refining current theoretical notions. Further, JoinInfer outperforms existing state-of-the-art inference engines (ACE, IJGP and libDAI) on some standard benchmark datasets by up to a factor of 630x. Finally, we propose a promising data-driven heuristic that extends JoinInfer to automatically tailor its parameters and/or switch to the traditional inference algorithms.