Beyond Message Passing: Neural Graph Pattern Machine
This addresses a fundamental limitation in graph learning for tasks such as social network analysis and molecular graph processing, though it is a novel method rather than a new paradigm.
The paper tackled the problem of graph neural networks struggling to capture key substructure patterns like triangles and rings, which limits expressiveness and long-range dependency modeling, by introducing the Neural Graph Pattern Machine (GPM) that learns directly from graph substructures, resulting in outperforming state-of-the-art baselines across four standard tasks.
Graph learning tasks often hinge on identifying key substructure patterns -- such as triadic closures in social networks or benzene rings in molecular graphs -- that underpin downstream performance. However, most existing graph neural networks (GNNs) rely on message passing, which aggregates local neighborhood information iteratively and struggles to explicitly capture such fundamental motifs, like triangles, k-cliques, and rings. This limitation hinders both expressiveness and long-range dependency modeling. In this paper, we introduce the Neural Graph Pattern Machine (GPM), a novel framework that bypasses message passing by learning directly from graph substructures. GPM efficiently extracts, encodes, and prioritizes task-relevant graph patterns, offering greater expressivity and improved ability to capture long-range dependencies. Empirical evaluations across four standard tasks -- node classification, link prediction, graph classification, and graph regression -- demonstrate that GPM outperforms state-of-the-art baselines. Further analysis reveals that GPM exhibits strong out-of-distribution generalization, desirable scalability, and enhanced interpretability. Code and datasets are available at: https://github.com/Zehong-Wang/GPM.