PreGIP: Watermarking the Pretraining of Graph Neural Networks for Deep Intellectual Property Protection

Abstract

Pretraining on Graph Neural Networks (GNNs) has shown great power in facilitating various downstream tasks. As pretraining generally requires huge amount of data and computational resources, the pretrained GNNs are high-value Intellectual Properties (IP) of the legitimate owner. However, adversaries may illegally copy and deploy the pretrained GNN models for their downstream tasks. Though initial efforts have been made to watermark GNN classifiers for IP protection, these methods require the target classification task for watermarking, and thus are not applicable to self-supervised pretraining of GNN models. Hence, in this work, we propose a novel framework named PreGIP to watermark the pretraining of GNN encoder for IP protection while maintain the high-quality of the embedding space. PreGIP incorporates a task-free watermarking loss to watermark the embedding space of pretrained GNN encoder. A finetuning-resistant watermark injection is further deployed. Theoretical analysis and extensive experiments show the effectiveness of {\method} in IP protection and maintaining high-performance for downstream tasks.

References in corpus (13)

• GCC: Graph Contrastive Coding for Graph Neural Network Pre-Training
• Certified Adversarial Robustness via Randomized Smoothing
• A Comprehensive Survey on Trustworthy Graph Neural Networks: Privacy, Robustness, Fairness, and Explainability
• Backdoor Pre-trained Models Can Transfer to All
• Unnoticeable Backdoor Attacks on Graph Neural Networks
• GraphMAE: Self-Supervised Masked Graph Autoencoders
• NRGNN: Learning a Label Noise-Resistant Graph Neural Network on Sparsely and Noisily Labeled Graphs
• Certified Neural Network Watermarks with Randomized Smoothing
• SSLGuard: A Watermarking Scheme for Self-supervised Learning Pre-trained Encoders
• Certifiably Robust Graph Contrastive Learning
• Watermarking Graph Neural Networks based on Backdoor Attacks
• Watermarking Pre-trained Encoders in Contrastive Learning
• A Unified Framework of Graph Information Bottleneck for Robustness and Membership Privacy