GraphDLG: Exploring Deep Leakage from Gradients in Federated Graph Learning

TL;DR

This paper introduces GraphDLG, a method to recover raw graph data from shared gradients in federated graph learning, revealing privacy vulnerabilities and outperforming existing approaches in reconstructing graph structures and node features.

Contribution

The paper provides the first theoretical analysis of gradient leakage in federated graph learning and proposes GraphDLG, a novel method for effective graph data reconstruction from gradients.

Findings

GraphDLG outperforms existing methods in reconstructing graph structures.

GraphDLG achieves over 5.46% improvement in node feature reconstruction (MSE).

GraphDLG achieves over 25.04% improvement in graph structure reconstruction (AUC).

Abstract

Federated graph learning (FGL) has recently emerged as a promising privacy-preserving paradigm that enables distributed graph learning across multiple data owners. A critical privacy concern in federated learning is whether an adversary can recover raw data from shared gradients, a vulnerability known as deep leakage from gradients (DLG). However, most prior studies on the DLG problem focused on image or text data, and it remains an open question whether graphs can be effectively recovered, particularly when the graph structure and node features are uniquely entangled in GNNs. In this work, we first theoretically analyze the components in FGL and derive a crucial insight: once the graph structure is recovered, node features can be obtained through a closed-form recursive rule. Building on this analysis, we propose GraphDLG, a novel approach to recover raw training graphs from shared gradients in FGL, which can utilize randomly generated graphs or client-side training graphs as auxiliaries to enhance recovery. Extensive experiments demonstrate that GraphDLG outperforms existing solutions by successfully decoupling the graph structure and node features, achieving improvements of over 5.46% (by MSE) for node feature reconstruction and over 25.04% (by AUC) for graph structure reconstruction.