Leap: Inductive Link Prediction via Learnable TopologyAugmentation
This work addresses the need for more expressive inductive link prediction methods in graph machine learning, which is crucial for real-world applications involving new nodes, though it appears incremental as it builds on existing GNN and MLP approaches.
The paper tackles the problem of inductive link prediction for new nodes in graphs, proposing LEAP, a method that uses learnable topology augmentation to incorporate structural and feature information, achieving improvements of up to 22% in AUC and 17% in average precision over state-of-the-art methods.
Link prediction is a crucial task in many downstream applications of graph machine learning. To this end, Graph Neural Network (GNN) is a widely used technique for link prediction, mainly in transductive settings, where the goal is to predict missing links between existing nodes. However, many real-life applications require an inductive setting that accommodates for new nodes, coming into an existing graph. Thus, recently inductive link prediction has attracted considerable attention, and a multi-layer perceptron (MLP) is the popular choice of most studies to learn node representations. However, these approaches have limited expressivity and do not fully capture the graph's structural signal. Therefore, in this work we propose LEAP, an inductive link prediction method based on LEArnable toPology augmentation. Unlike previous methods, LEAP models the inductive bias from both the structure and node features, and hence is more expressive. To the best of our knowledge, this is the first attempt to provide structural contexts for new nodes via learnable augmentation in inductive settings. Extensive experiments on seven real-world homogeneous and heterogeneous graphs demonstrates that LEAP significantly surpasses SOTA methods. The improvements are up to 22\% and 17\% in terms of AUC and average precision, respectively. The code and datasets are available on GitHub (https://github.com/AhmedESamy/LEAP/)