Beyond Observed Connections : Link Injection

TL;DR

This paper introduces link injection, a novel end-to-end method that enables graph models to discover and utilize unobserved connections, improving performance on node classification and link prediction tasks.

Contribution

The paper presents a new link injection layer that allows differentiable graph models to incorporate unseen connections, enhancing their ability to exploit hidden relationships.

Findings

Improved performance on node classification tasks.

Enhanced link prediction accuracy.

Effective exploitation of unseen connections.

Abstract

In this paper, we proposed the \textit{link injection}, a novel method that helps any differentiable graph machine learning models to go beyond observed connections from the input data in an end-to-end learning fashion. It finds out (weak) connections in favor of the current task that is not present in the input data via a parametric link injection layer. We evaluate our method on both node classification and link prediction tasks using a series of state-of-the-art graph convolution networks. Results show that the link injection helps a variety of models to achieve better performances on both applications. Further empirical analysis shows a great potential of this method in efficiently exploiting unseen connections from the injected links.