Adversarial Robustness in Graph Neural Networks: A Hamiltonian Approach
Authors:
- Qiyu KangQKQiyu Kang
- Rui SheRSRui She
Preprint
Abstract
Graph neural networks (GNNs) are vulnerable to adversarial perturbations, including those that affect both node features and graph topology. This paper investigates GNNs derived from diverse neural flows, concentrating on their connection to various stability notions such as BIBO stability, Lyapunov stability, structural stability, and conservative stability. We argue that Lyapunov stability, despite its common use, does not necessarily ensure adversarial robustness. Inspired by physics principles, we advocate for the use of conservative Hamiltonian neural flows to construct GNNs that are robust to adversarial attacks. The adversarial robustness of different neural flow GNNs is empirically compared on several benchmark datasets under a variety of adversarial attacks. Extensive numerical experiments demonstrate that GNNs leveraging conservative Hamiltonian flows with Lyapunov stability substantially improve robustness against adversarial perturbations. The implementation code of experiments is available at https://github.com/zknus/NeurIPS-2023-HANG-Robustness.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Image Patch-Matching With Graph-Based Learning in Street ScenesIEEE Transactions on Image Processing, 2023
- Single-node attacks for fooling graph neural networksNeurocomputing, 2022
- Physics-enhanced neural networks learn order and chaosPhysical Review E, 2020
- Scalable attack on graph data by injecting vicious nodesData Mining and Knowledge Discovery, 2020
- Graph embedding and unsupervised learning predict genomic sub-compartments from HiC chromatin interaction dataNature Communications, 2020
- Graph embedding on biomedical networks: methods, applications and evaluationsBioinformatics, 2019
- Hiding individuals and communities in a social networkNature Human Behaviour, 2018
- Stable architectures for deep neural networksInverse Problems, 2017
- Differential Equations, Dynamical Systems, and an Introduction to Chaos2013
- Introduction to the Modern Theory of Dynamical Systems1995