Extracting Privacy-Preserving Subgraphs in Federated Graph Learning using Information Bottleneck

Zhang, Chenhan1; Wang, Weiqi1; Yu, James J. Q.2; Yu, Shui1

10.1145/3579856.3595791

2023

18th ACM ASIA Conference on Computer and Communications Security (ASIA CCS)

PROCEEDINGS OF THE 2023 ACM ASIA CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY, ASIA CCS 2023

JUL 10-14, 2023

null,Melbourne,AUSTRALIA

1601 Broadway, 10th Floor, NEW YORK, NY, UNITED STATES

ASSOC COMPUTING MACHINERY

As graphs are getting larger and larger, federated graph learning (FGL) is increasingly adopted, which can train graph neural networks (GNNs) on distributed graph data. However, the privacy of graph data in FGL systems is an inevitable concern due to multiparty participation. Recent studies indicated that the gradient leakage of trained GNN can be used to infer private graph data information utilizing model inversion attacks (MIA). Moreover, the central server can legitimately access the localGNNgradients, which makes MIA difficult to counter if the attacker is at the central server. In this paper, we first identify a realistic crowdsourcing-based FGL scenario where MIA from the central server towards clients' subgraph structures is a nonnegligible threat. Then, we propose a defense scheme, Subgraph-Out-of-Subgraph (SOS), to mitigate such MIA and meanwhile, maintain the prediction accuracy. We leverage the information bottleneck (IB) principle to extract task-relevant subgraphs out of the clients' original subgraphs. The extracted IB-subgraphs are used for local GNN training and the local model updates will have less information about the original subgraphs, which renders the MIA harder to infer the original subgraph structure. Particularly, we devise a novel neural network-powered approach to overcome the intractability of graph data's mutual information estimation in IB optimization. Additionally, we design a subgraph generation algorithm for finally yielding reasonable IB-subgraphs from the optimization results. Extensive experiments demonstrate the efficacy of the proposed scheme, the FGL system trained on IB-subgraphs is more robust against MIA attacks with minuscule accuracy loss.

Graph Neural Networks Federated Learning Model Inversion Attack Information Bottleneck

Others

English

CPCI-S

Australian Research Council (ARC)["LP190100676","DP200101374"]

Computer Science ; Mathematics ; Telecommunications

Computer Science, Artificial Intelligence ; Mathematics, Applied ; Telecommunications

WOS:001053857900010

Web of Science

1.Univ Technol Sydney, Sydney, Australia
2.Southern Univ Sci & Technol, Shenzhen, Peoples R China

Zhang, Chenhan,Wang, Weiqi,Yu, James J. Q.,et al. Extracting Privacy-Preserving Subgraphs in Federated Graph Learning using Information Bottleneck[C]. 1601 Broadway, 10th Floor, NEW YORK, NY, UNITED STATES:ASSOC COMPUTING MACHINERY,2023.