Meta-Inductive Node Classification across Graphs

TL;DR

This paper introduces MI-GNN, a meta-learning framework for inductive node classification across different graphs, which adapts to inter-graph differences and improves performance over traditional GNNs.

Contribution

The paper proposes a novel meta-inductive framework, MI-GNN, that learns to adapt models to new graphs by capturing graph and task-level differences.

Findings

MI-GNN outperforms existing methods on five real-world graph datasets.

The dual adaptation mechanism effectively handles inter-graph variability.

Meta-learning improves generalization to unseen graphs.

Abstract

Semi-supervised node classification on graphs is an important research problem, with many real-world applications in information retrieval such as content classification on a social network and query intent classification on an e-commerce query graph. While traditional approaches are largely transductive, recent graph neural networks (GNNs) integrate node features with network structures, thus enabling inductive node classification models that can be applied to new nodes or even new graphs in the same feature space. However, inter-graph differences still exist across graphs within the same domain. Thus, training just one global model (e.g., a state-of-the-art GNN) to handle all new graphs, whilst ignoring the inter-graph differences, can lead to suboptimal performance. In this paper, we study the problem of inductive node classification across graphs. Unlike existing one-model-fits-all approaches, we propose a novel meta-inductive framework called MI-GNN to customize the inductive model to each graph under a meta-learning paradigm. That is, MI-GNN does not directly learn an inductive model; it learns the general knowledge of how to train a model for semi-supervised node classification on new graphs. To cope with the differences across graphs, MI-GNN employs a dual adaptation mechanism at both the graph and task levels. More specifically, we learn a graph prior to adapt for the graph-level differences, and a task prior to adapt for the task-level differences conditioned on a graph. Extensive experiments on five real-world graph collections demonstrate the effectiveness of our proposed model.