Bipartite Graph Embedding via Mutual Information Maximization

TL;DR

This paper introduces BiGI, a bipartite graph embedding method that maximizes mutual information between local subgraph representations and global prototypes to better capture global graph properties like community structures and long-range dependencies.

Contribution

BiGI is the first bipartite graph embedding approach that explicitly models global properties using a local-global infomax objective, improving over local-only methods.

Findings

BiGI outperforms state-of-the-art baselines on recommendation and link prediction tasks.

BiGI effectively captures global community structures and long-range dependencies.

Extensive experiments confirm the superiority of the mutual information maximization approach.

Abstract

Bipartite graph embedding has recently attracted much attention due to the fact that bipartite graphs are widely used in various application domains. Most previous methods, which adopt random walk-based or reconstruction-based objectives, are typically effective to learn local graph structures. However, the global properties of bipartite graph, including community structures of homogeneous nodes and long-range dependencies of heterogeneous nodes, are not well preserved. In this paper, we propose a bipartite graph embedding called BiGI to capture such global properties by introducing a novel local-global infomax objective. Specifically, BiGI first generates a global representation which is composed of two prototype representations. BiGI then encodes sampled edges as local representations via the proposed subgraph-level attention mechanism. Through maximizing the mutual information between local and global representations, BiGI enables nodes in bipartite graph to be globally relevant. Our model is evaluated on various benchmark datasets for the tasks of top-K recommendation and link prediction. Extensive experiments demonstrate that BiGI achieves consistent and significant improvements over state-of-the-art baselines. Detailed analyses verify the high effectiveness of modeling the global properties of bipartite graph.