Scaling Inference for Markov Logic with a Task-Decomposition Approach

TL;DR

This paper introduces Felix, a task-decomposition approach using Lagrangian relaxation to enable scalable, joint inference in Markov Logic Networks, significantly improving efficiency on large knowledge bases.

Contribution

The paper presents Felix, a novel architecture that decomposes MLN inference tasks, leveraging RDBMS and specialized algorithms for scalability and efficiency.

Findings

Felix outperforms prior MLN inference methods in scalability and efficiency.

Successfully applied to a knowledge base from 1.8 million documents.

Achieves state-of-the-art inference quality on large-scale data.

Abstract

Motivated by applications in large-scale knowledge base construction, we study the problem of scaling up a sophisticated statistical inference framework called Markov Logic Networks (MLNs). Our approach, Felix, uses the idea of Lagrangian relaxation from mathematical programming to decompose a program into smaller tasks while preserving the joint-inference property of the original MLN. The advantage is that we can use highly scalable specialized algorithms for common tasks such as classification and coreference. We propose an architecture to support Lagrangian relaxation in an RDBMS which we show enables scalable joint inference for MLNs. We empirically validate that Felix is significantly more scalable and efficient than prior approaches to MLN inference by constructing a knowledge base from 1.8M documents as part of the TAC challenge. We show that Felix scales and achieves state-of-the-art quality numbers. In contrast, prior approaches do not scale even to a subset of the corpus that is three orders of magnitude smaller.