Fast Parallel Exact Inference on Bayesian Networks: Poster

TL;DR

This paper introduces Fast-BNI, a parallel algorithm for exact inference in Bayesian networks that significantly improves efficiency on multi-core CPUs by combining different levels of parallelism and optimizing bottleneck operations.

Contribution

The paper presents Fast-BNI, a novel hybrid parallel exact inference method for Bayesian networks optimized for multi-core CPUs, with open-source implementation.

Findings

Achieves faster exact inference on complex Bayesian networks.

Effectively combines coarse- and fine-grained parallelism.

Source code available for public use.

Abstract

Bayesian networks (BNs) are attractive, because they are graphical and interpretable machine learning models. However, exact inference on BNs is time-consuming, especially for complex problems. To improve the efficiency, we propose a fast BN exact inference solution named Fast-BNI on multi-core CPUs. Fast-BNI enhances the efficiency of exact inference through hybrid parallelism that tightly integrates coarse- and fine-grained parallelism. We also propose techniques to further simplify the bottleneck operations of BN exact inference. Fast-BNI source code is freely available at https://github.com/jjiantong/FastBN.