How Neural Processes Improve Graph Link Prediction
This addresses the inductive link prediction problem for real-world graph applications, representing an incremental improvement over existing methods.
The paper tackles the challenging problem of inductive link prediction on graphs, where only a subset of nodes and links are available during training, and proposes a meta-learning approach with graph neural networks that achieves stronger performance compared to state-of-the-art models and generalizes well with small subgraphs.
Link prediction is a fundamental problem in graph data analysis. While most of the literature focuses on transductive link prediction that requires all the graph nodes and majority of links in training, inductive link prediction, which only uses a proportion of the nodes and their links in training, is a more challenging problem in various real-world applications. In this paper, we propose a meta-learning approach with graph neural networks for link prediction: Neural Processes for Graph Neural Networks (NPGNN), which can perform both transductive and inductive learning tasks and adapt to patterns in a large new graph after training with a small subgraph. Experiments on real-world graphs are conducted to validate our model, where the results suggest that the proposed method achieves stronger performance compared to other state-of-the-art models, and meanwhile generalizes well when training on a small subgraph.