Federated Cross-Client Subgraph Pattern Detection

2026-05-07

AI Summary

This work addresses the limited representational capacity of existing federated graph learning methods, which struggle to capture cross-client subgraph patterns due to each client’s access being restricted to a local subgraph without global structural information. The study formulates this challenge as a structural observability problem and introduces a layer-wise forward embedding exchange framework that approximates the expressiveness of centralized graph neural networks (GNNs) by synchronizing intermediate node representations in real time, while preserving the privacy of raw features and labels. By integrating parameter aggregation with representation alignment under an expanded subgraph assumption, the proposed approach effectively overcomes the locality constraints inherent in conventional federated GNNs. Experiments on synthetic directed multigraphs featuring cycles, bicliques, and disassortative structures demonstrate that the method substantially narrows the representation gap with centralized models.

Abstract

Subgraph pattern detection aims to uncover complex interaction structures in graphs. However, state-of-the-art graph neural network (GNN)-based solutions assume centralized access to the entire graph. When graphs are instead distributed across multiple parties, client-local GNN computations diverge from those of a centralized model, resulting in a representation-equivalence gap. We formalize this as a structural observability problem, where subgraph patterns crossing partition boundaries become locally unidentifiable. To bridge this gap, we propose a per-step, layer-wise embedding exchange framework in which clients synchronize intermediate node representations at each layer of the forward pass, without exposing raw features or labels. Under an extended-subgraph assumption and shared model parameters across clients, this framework recovers the same node representations as a centralized GNN over the full graph. Experiments on synthetic directed multigraphs with cycles, bicliques, and scatter-gather patterns show that embedding exchange and federated parameter aggregation are complementary rather than interchangeable: their combination recovers most of the representation gap, provided exchanged embeddings are fresh per-step rather than stale per-epoch.

Problem

• federated learning
• subgraph pattern detection
• graph neural networks
• structural observability
• representation equivalence

Innovation

• federated learning
• subgraph pattern detection
• graph neural networks
• embedding exchange
• structural observability