SLAPS: Self-Supervision Improves Structure Learning for Graph Neural Networks

TL;DR

SLAPS introduces a self-supervised method for jointly learning graph structures and GNN parameters, significantly improving structure inference and scalability on large graphs.

Contribution

It proposes a novel self-supervised approach for inferring task-specific graph structures, enhancing GNN performance without requiring explicit structural supervision.

Findings

SLAPS scales to graphs with hundreds of thousands of nodes

Outperforms existing models on benchmark datasets

Provides more supervision for structure learning through self-supervision

Abstract

Graph neural networks (GNNs) work well when the graph structure is provided. However, this structure may not always be available in real-world applications. One solution to this problem is to infer a task-specific latent structure and then apply a GNN to the inferred graph. Unfortunately, the space of possible graph structures grows super-exponentially with the number of nodes and so the task-specific supervision may be insufficient for learning both the structure and the GNN parameters. In this work, we propose the Simultaneous Learning of Adjacency and GNN Parameters with Self-supervision, or SLAPS, a method that provides more supervision for inferring a graph structure through self-supervision. A comprehensive experimental study demonstrates that SLAPS scales to large graphs with hundreds of thousands of nodes and outperforms several models that have been proposed to learn a task-specific graph structure on established benchmarks.