Inductive Subgraph Embedding for Link Prediction

TL;DR

This paper introduces SCLRL, a self-supervised framework that leverages neighborhood subgraphs for effective and scalable link representation learning, outperforming existing methods in accuracy and efficiency.

Contribution

It proposes a novel subgraph contrastive learning approach for link embedding that captures intrinsic link features and improves scalability on large graphs.

Findings

SCLRL outperforms existing methods on benchmark datasets.

SCLRL demonstrates significant efficiency in training speed and memory usage.

Subgraph-level contrastive discrimination effectively captures link features.

Abstract

Graph representation learning (GRL) has emerged as a powerful technique for solving graph analytics tasks. It can effectively convert discrete graph data into a low-dimensional space where the graph structural information and graph properties are maximumly preserved. While there is rich literature on node and whole-graph representation learning, GRL for link is relatively less studied and less understood. One common practice in previous works is to generate link representations by directly aggregating the representations of their incident nodes, which is not capable of capturing effective link features. Moreover, common GRL methods usually rely on full-graph training, suffering from poor scalability and high resource consumption on large-scale graphs. In this paper, we design Subgraph Contrastive Link Representation Learning (SCLRL) -- a self-supervised link embedding framework, which utilizes the strong correlation between central links and their neighborhood subgraphs to characterize links. We extract the "link-centric induced subgraphs" as input, with a subgraph-level contrastive discrimination as pretext task, to learn the intrinsic and structural link features via subgraph mini-batch training. Extensive experiments conducted on five datasets demonstrate that SCLRL has significant performance advantages in link representation learning on benchmark datasets and prominent efficiency advantages in terms of training speed and memory consumption on large-scale graphs, when compared with existing link representation learning methods.