Mixture of Link Predictors on Graphs
This addresses the challenge of suboptimal performance in graph link prediction for researchers and practitioners by adapting to diverse pairwise information needs, representing an incremental advance over existing GNN methods.
The paper tackles the problem of link prediction in graphs by proposing Link-MoE, a mixture of experts model that selects different GNN experts for each node pair based on varied pairwise information, achieving relative improvements of 18.71% on MRR for Pubmed and 9.59% on Hits@100 for ogbl-ppa compared to baselines.
Link prediction, which aims to forecast unseen connections in graphs, is a fundamental task in graph machine learning. Heuristic methods, leveraging a range of different pairwise measures such as common neighbors and shortest paths, often rival the performance of vanilla Graph Neural Networks (GNNs). Therefore, recent advancements in GNNs for link prediction (GNN4LP) have primarily focused on integrating one or a few types of pairwise information. In this work, we reveal that different node pairs within the same dataset necessitate varied pairwise information for accurate prediction and models that only apply the same pairwise information uniformly could achieve suboptimal performance. As a result, we propose a simple mixture of experts model Link-MoE for link prediction. Link-MoE utilizes various GNNs as experts and strategically selects the appropriate expert for each node pair based on various types of pairwise information. Experimental results across diverse real-world datasets demonstrate substantial performance improvement from Link-MoE. Notably, Link-MoE achieves a relative improvement of 18.71\% on the MRR metric for the Pubmed dataset and 9.59\% on the Hits@100 metric for the ogbl-ppa dataset, compared to the best baselines.