CryptGNN: Enabling Secure Inference for Graph Neural Networks

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present CryptGNN, a secure and effective inference solution for third-party graph neural network (GNN) models in the cloud, which are accessed by clients as ML as a service (MLaaS). The main novelty of CryptGNN is its secure message passing and feature transformation layers using distributed secure multi-party computation (SMPC) techniques. CryptGNN protects the client's input data and graph structure from the cloud provider and the third-party model owner, and it protects the model parameters from the cloud provider and the clients. CryptGNN works with any number of SMPC parties, does not require a trusted server, and is provably secure even if P − 1 out of P parties in the cloud collude. Theoretical analysis and empirical experiments demonstrate the security and efficiency of CryptGNN.

Original languageEnglish (US)
Title of host publicationCCS 2025 - Proceedings of the 2025 ACM SIGSAC Conference on Computer and Communications Security
PublisherAssociation for Computing Machinery, Inc
Pages291-305
Number of pages15
ISBN (Electronic)9798400715259
DOIs
StatePublished - Nov 22 2025
Event32nd ACM SIGSAC Conference on Computer and Communications Security, CCS 2025 - Taipei, Taiwan, Province of China
Duration: Oct 13 2025Oct 17 2025

Publication series

NameCCS 2025 - Proceedings of the 2025 ACM SIGSAC Conference on Computer and Communications Security

Conference

Conference32nd ACM SIGSAC Conference on Computer and Communications Security, CCS 2025
Country/TerritoryTaiwan, Province of China
CityTaipei
Period10/13/2510/17/25

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Networks and Communications
  • Computer Science Applications

Keywords

  • Graph Neural Networks
  • Machine Learning as a Service
  • Secure Multi-party Computation

Fingerprint

Dive into the research topics of 'CryptGNN: Enabling Secure Inference for Graph Neural Networks'. Together they form a unique fingerprint.

Cite this