Differentiable Learning of Logical Rules for Knowledge Base Reasoning

TL;DR

This paper introduces Neural Logic Programming, a differentiable framework for learning probabilistic logical rules for knowledge base reasoning, enabling end-to-end learning of rule structure and parameters, and demonstrating superior performance on benchmark datasets.

Contribution

It presents a novel neural framework that combines structure and parameter learning of logical rules in an end-to-end differentiable manner for knowledge base reasoning.

Findings

Outperforms prior methods on Freebase and WikiMovies datasets.

Successfully learns probabilistic logical rules in a differentiable setting.

Demonstrates effective end-to-end training of rule structure and parameters.

Abstract

We study the problem of learning probabilistic first-order logical rules for knowledge base reasoning. This learning problem is difficult because it requires learning the parameters in a continuous space as well as the structure in a discrete space. We propose a framework, Neural Logic Programming, that combines the parameter and structure learning of first-order logical rules in an end-to-end differentiable model. This approach is inspired by a recently-developed differentiable logic called TensorLog, where inference tasks can be compiled into sequences of differentiable operations. We design a neural controller system that learns to compose these operations. Empirically, our method outperforms prior work on multiple knowledge base benchmark datasets, including Freebase and WikiMovies.