Graph Representation Learning via Graphical Mutual Information Maximization

TL;DR

This paper introduces Graphical Mutual Information (GMI), a novel unsupervised method for learning graph representations by maximizing mutual information between input graphs and their embeddings, improving performance on node classification and link prediction tasks.

Contribution

It proposes GMI, a new graph mutual information measure invariant to graph isomorphisms, and develops an unsupervised learning model that outperforms existing methods.

Findings

Outperforms state-of-the-art unsupervised graph learning methods.

Sometimes exceeds supervised learning performance.

Effective on both transductive and inductive tasks.

Abstract

The richness in the content of various information networks such as social networks and communication networks provides the unprecedented potential for learning high-quality expressive representations without external supervision. This paper investigates how to preserve and extract the abundant information from graph-structured data into embedding space in an unsupervised manner. To this end, we propose a novel concept, Graphical Mutual Information (GMI), to measure the correlation between input graphs and high-level hidden representations. GMI generalizes the idea of conventional mutual information computations from vector space to the graph domain where measuring mutual information from two aspects of node features and topological structure is indispensable. GMI exhibits several benefits: First, it is invariant to the isomorphic transformation of input graphs---an inevitable constraint in many existing graph representation learning algorithms; Besides, it can be efficiently estimated and maximized by current mutual information estimation methods such as MINE; Finally, our theoretical analysis confirms its correctness and rationality. With the aid of GMI, we develop an unsupervised learning model trained by maximizing GMI between the input and output of a graph neural encoder. Considerable experiments on transductive as well as inductive node classification and link prediction demonstrate that our method outperforms state-of-the-art unsupervised counterparts, and even sometimes exceeds the performance of supervised ones.