Graph Transductive Defense: a Two-Stage Defense for Graph Membership Inference Attacks

TL;DR

This paper introduces a novel two-stage defense mechanism called Graph Transductive Defense (GTD) that effectively mitigates membership inference attacks on graph neural networks in transductive learning settings, improving privacy and utility.

Contribution

The paper proposes GTD, a tailored defense strategy for GNNs in transductive learning, combining train-test alternate training and flattening to reduce attack success and enhance model utility.

Findings

GTD reduces attack AUROC by 9.42% on average.

GTD increases utility performance by 18.08%.

The method integrates seamlessly with minimal overhead.

Abstract

Graph neural networks (GNNs) have become instrumental in diverse real-world applications, offering powerful graph learning capabilities for tasks such as social networks and medical data analysis. Despite their successes, GNNs are vulnerable to adversarial attacks, including membership inference attacks (MIA), which threaten privacy by identifying whether a record was part of the model's training data. While existing research has explored MIA in GNNs under graph inductive learning settings, the more common and challenging graph transductive learning setting remains understudied in this context. This paper addresses this gap and proposes an effective two-stage defense, Graph Transductive Defense (GTD), tailored to graph transductive learning characteristics. The gist of our approach is a combination of a train-test alternate training schedule and flattening strategy, which successfully reduces the difference between the training and testing loss distributions. Extensive empirical results demonstrate the superior performance of our method (a decrease in attack AUROC by $9.42\%$ and an increase in utility performance by $18.08\%$ on average compared to LBP), highlighting its potential for seamless integration into various classification models with minimal overhead.