Encrypted Inference for Graph Neural Networks

 

Speaker: Associate Professor Xingliang Yuan, The University of Melbourne

 

Abstract: Graph Neural Networks (GNNs) become a core offering of MLaaS platforms. Protecting the intellectual property (IP) of model owners and the confidentiality of underlying graph data is deemed as a critical necessity. In particular, GNN models trained on proprietary data are considered as sensitive IP, whilst graph data often contains private information. Both of them demand stringent protection when being deployed in untrusted MLaaS. In this talk, I will present our recent work named OblivGNN. It is designed to efficiently enable encrypted inference over encrypted graph data and GNN models. OblivGNN is the first to support both transductive and inductive settings of GNNs in the encrypted domain. It ensures the confidentiality of graphs and models, and achieves a strong security property "data obliviousness" during encrypted inference. I will conclude with the open problems in this area and share our in-progress work that attempts to scale encrypted inference and training over large graphs. 

Bio: Dr Xingliang Yuan is an Associate Professor in the School of Computing and Information Systems at the University of Melbourne, and an ARC Future Fellow. Previously, he was a Lecturer, then Senior Lecturer at the Faculty of IT, Monash University (2017–2024). His research focuses on designing secure networked systems and protocols to protect digital assets in untrusted environments. His research has been supported by ARC, CSIRO, the Australian Departments of Home Affairs, and Health and Aged Care, and Oceania Cyber Security Centre. Dr. Yuan's work is regularly published in major computer security and networked system venues. His external and internal contributions earned him several honours, including the Dean's Award for Excellence in Research for an ECR (2020), the Faculty Teaching Excellence Award (2021), the Excellence in Engagement Award at UniMelb (2024), the Best Paper Award at ESORICS (2021), and an Honourable Mention Paper Award at USENIX Security (2025). Dr. Yuan serves on the editorial board of IEEE TDSC and IEEE TSC (Area Editor in Security, Privacy, and Trust), and as a General Chair for ICDCS'27 and RAID'25, a PC Chair for Lamps@CCS'24, SecTL@AsiaCCS'23, and NSS'22, and a Track Chair for ICDCS'24. He also received the Notable TPC award recently at USENIX Security (2025).