mlsanityData to 14 Oct 2026

Graph in the Vault: Protecting Edge GNN Inference with Trusted Execution Environment

arXiv 2502.15012 ↗First posted 20 Feb 2025cs.CR cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Wide deployment of machine learning models on edge devices has rendered the model intellectual property (IP) and data privacy vulnerable. We propose GNNVault, the first secure Graph Neural Network (GNN) deployment strategy based on Trusted Execution Environment (TEE). GNNVault follows the design of 'partition-before-training' and includes a private GNN rectifier to complement with a public backbone model. This way, both critical GNN model parameters and the private graph used during inference are protected within secure TEE compartments. Real-world implementations with Intel SGX demonstrate that GNNVault safeguards GNN inference against state-of-the-art link stealing attacks with negligible accuracy degradation (<2%).

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 20 of its 33 references that are in the map. First posted on arXiv 20 Feb 2025.

Builds on

20 of 33 references.

References · 20

  1. 2018How Powerful are Graph Neural Networks?cited by 4,393
  2. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,281
  3. 2018Pitfalls of Graph Neural Network Evaluationcited by 1,007
  4. 2016Revisiting Semi-Supervised Learning with Graph Embeddingscited by 976
  5. 2020Graph Neural Networks in Recommender Systems: A Surveycited by 494
  6. 2017Deep Gaussian Embedding of Graphs: Unsupervised Inductive Learning via Rankingcited by 335
  7. 2018Session-based Recommendation with Graph Neural Networkscited by 312
  8. 2019AliGraph: A Comprehensive Graph Neural Network Platformcited by 153
  9. 2020Stealing Links from Graph Neural Networkscited by 115
  10. 2019Meta-GNN: On Few-shot Node Classification in Graph Meta-learningcited by 105
  11. 2019On-Device Machine Learning: An Algorithms and Learning Theory Perspectivecited by 63
  12. 2020DarkneTZ: Towards Model Privacy at the Edge using Trusted Execution Environmentscited by 58
8 more
  1. 2021Non-Transferable Learning: A New Approach for Model Ownership Verification and Applicability Authorizationcited by 52
  2. 2019Overlapping Community Detection with Graph Neural Networkscited by 45
  3. 2020Branchy-GNN: a Device-Edge Co-Inference Framework for Efficient Point Cloud Processingcited by 21
  4. 2023No Privacy Left Outside: On the (In-)Security of TEE-Shielded DNN Partition for On-Device MLcited by 21
  5. 2022GNN at the Edge: Cost-Efficient Graph Neural Network Processing over Distributed Edge Serverscited by 11
  6. 2023MirrorNet: A TEE-Friendly Framework for Secure On-device DNN Inferencecited by 9
  7. 2019BAYHENN: Combining Bayesian Deep Learning and Homomorphic Encryption for Secure DNN Inferencecited by 5
  8. 2023VertexSerum: Poisoning Graph Neural Networks for Link Inferencecited by 5

Built on by

Not cited yet.