Efficient Link Prediction via GNN Layers Induced by Negative Sampling

Graph neural networks (GNNs) for link prediction can loosely be divided into two broad categories. First, node-wise architectures pre-compute individual embeddings for each node that are later combined by a simple decoder to make predictions. While extremely efficient at inference time, model expres...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on knowledge and data engineering 2025-01, Vol.37 (1), p.253-264
Main Authors: Wang, Yuxin, Hu, Xiannian, Gan, Quan, Huang, Xuanjing, Qiu, Xipeng, Wipf, David
Format: Article
Language:English
Subjects:
Accuracy
Artificial intelligence (AI)
Computational modeling
Computer architecture
Convergence
Costs
Decoding
Graph neural networks
graph neural networks (GNN)
link prediction
machine learning
Optimization
Predictive models
Training
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Graph neural networks (GNNs) for link prediction can loosely be divided into two broad categories. First, node-wise architectures pre-compute individual embeddings for each node that are later combined by a simple decoder to make predictions. While extremely efficient at inference time, model expressiveness is limited such that isomorphic nodes contributing to candidate edges may not be distinguishable, compromising accuracy. In contrast, edge-wise methods rely on the formation of edge-specific subgraph embeddings to enrich the representation of pair-wise relationships, disambiguating isomorphic nodes to improve accuracy, but with increased model complexity. To better navigate this trade-off, we propose a novel GNN architecture whereby the forward pass explicitly depends on both positive (as is typical) and negative (unique to our approach) edges to inform more flexible, yet still cheap node-wise embeddings. This is achieved by recasting the embeddings themselves as minimizers of a forward-pass-specific energy function that favors separation of positive and negative samples. Notably, this energy is distinct from the actual training loss shared by most existing link prediction models, where contrastive pairs only influence the backward pass . As demonstrated by extensive empirical evaluations, the resulting architecture retains the inference speed of node-wise models, while producing competitive accuracy with edge-wise alternatives.
ISSN:1041-4347
1558-2191
DOI:10.1109/TKDE.2024.3481015