Graph Matching Networks for Learning the Similarity of Graph Structured Objects

TL;DR

This paper introduces Graph Matching Networks that leverage Graph Neural Networks and a novel attention mechanism to effectively compute similarities between graph-structured objects, outperforming domain-specific baselines.

Contribution

It presents a new graph matching model with cross-graph attention, enabling improved similarity reasoning on graph data.

Findings

Effective graph embeddings for similarity reasoning

Outperforms domain-specific baseline systems

Applicable to control-flow-graph based function similarity search

Abstract

This paper addresses the challenging problem of retrieval and matching of graph structured objects, and makes two key contributions. First, we demonstrate how Graph Neural Networks (GNN), which have emerged as an effective model for various supervised prediction problems defined on structured data, can be trained to produce embedding of graphs in vector spaces that enables efficient similarity reasoning. Second, we propose a novel Graph Matching Network model that, given a pair of graphs as input, computes a similarity score between them by jointly reasoning on the pair through a new cross-graph attention-based matching mechanism. We demonstrate the effectiveness of our models on different domains including the challenging problem of control-flow-graph based function similarity search that plays an important role in the detection of vulnerabilities in software systems. The experimental analysis demonstrates that our models are not only able to exploit structure in the context of similarity learning but they can also outperform domain-specific baseline systems that have been carefully hand-engineered for these problems.