DRUM: End-To-End Differentiable Rule Mining On Knowledge Graphs

TL;DR

DRUM is a scalable, differentiable method for mining probabilistic logical rules from knowledge graphs, enabling inductive link prediction and interpretability, outperforming existing methods on benchmarks.

Contribution

It introduces a novel differentiable approach using RNNs for rule mining that handles unseen entities and provides interpretable probabilistic rules.

Findings

DRUM outperforms existing rule mining methods on benchmark datasets.

It effectively handles inductive link prediction with unseen entities.

The method offers interpretable probabilistic rules for knowledge graphs.

Abstract

In this paper, we study the problem of learning probabilistic logical rules for inductive and interpretable link prediction. Despite the importance of inductive link prediction, most previous works focused on transductive link prediction and cannot manage previously unseen entities. Moreover, they are black-box models that are not easily explainable for humans. We propose DRUM, a scalable and differentiable approach for mining first-order logical rules from knowledge graphs which resolves these problems. We motivate our method by making a connection between learning confidence scores for each rule and low-rank tensor approximation. DRUM uses bidirectional RNNs to share useful information across the tasks of learning rules for different relations. We also empirically demonstrate the efficiency of DRUM over existing rule mining methods for inductive link prediction on a variety of benchmark datasets.