Beyond the Grounding Bottleneck: Datalog Techniques for Inference in Probabilistic Logic Programs (Technical Report)

TL;DR

This paper introduces a novel Datalog-based inference method for probabilistic logic programs that overcomes the grounding bottleneck, enabling efficient query answering over large knowledge graphs and providing fast approximate results.

Contribution

It presents an innovative approach combining Datalog techniques with knowledge compilation to improve inference efficiency in probabilistic logic programming.

Findings

Eliminates the grounding bottleneck in probabilistic logic inference.

Achieves faster approximate inference on classical benchmarks.

Enables query answering over large knowledge graphs.

Abstract

State-of-the-art inference approaches in probabilistic logic programming typically start by computing the relevant ground program with respect to the queries of interest, and then use this program for probabilistic inference using knowledge compilation and weighted model counting. We propose an alternative approach that uses efficient Datalog techniques to integrate knowledge compilation with forward reasoning with a non-ground program. This effectively eliminates the grounding bottleneck that so far has prohibited the application of probabilistic logic programming in query answering scenarios over knowledge graphs, while also providing fast approximations on classical benchmarks in the field.