Conference proceeding
Topology Attack and Defense for Graph Neural Networks: An Optimization Perspective
PROCEEDINGS OF THE TWENTY-EIGHTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, pp.3961-3967
01 Jan 2019
Abstract
Graph neural networks (GNNs) which apply the deep neural networks to graph data have achieved significant performance for the task of semi-supervised node classification. However, only few work has addressed the adversarial robust-ness of GNNs. In this paper, we first present a novel gradient-based attack method that facilitates the difficulty of tackling discrete graph data. When comparing to current adversarial attacks on GNNs, the results show that by only perturbing a small number of edge perturbations, including addition and deletion, our optimization-based attack can lead to a noticeable decrease in classification performance. Moreover, leveraging our gradient-based attack, we propose the first optimization-based adversarial training for GNNs. Our method yields higher robustness against both different gradient based and greedy attack methods without sacrificing classification accuracy on original graph.
Metrics
4 Record Views
Details
- Title
- Topology Attack and Defense for Graph Neural Networks: An Optimization Perspective
- Creators
- Kaidi Xu - Northeastern UniversityHongge Chen - MIT, Elect Engn & Comp Sci, 77 Massachusetts Ave, Cambridge, MA 02139 USASijia Liu - IBMPin-Yu Chen - IBM Res, MIT IBM Watson AI Lab, Armonk, NY USATsui-Wei Weng - MIT, Elect Engn & Comp Sci, 77 Massachusetts Ave, Cambridge, MA 02139 USAMingyi Hong - University of MinnesotaXue Lin - Northeastern Univ, Elect & Comp Engn, Boston, MA 02115 USA
- Contributors
- S Kraus (Editor)
- Publication Details
- PROCEEDINGS OF THE TWENTY-EIGHTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, pp.3961-3967
- Conference
- TWENTY-EIGHTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 28th
- Publisher
- Ijcai-Int Joint Conf Artif Intell
- Number of pages
- 7
- Grant note
- FA8750-18-2-0058 / Air Force Research Laboratory MIT-IBM Watson AI Lab; International Business Machines (IBM)
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Computer Science (Computing)
- Identifiers
- 991021871354404721
InCites Highlights
These are selected metrics from InCites Benchmarking & Analytics tool, related to this output
- Collaboration types
- Industry collaboration
- Domestic collaboration
- Web of Science research areas
- Computer Science, Artificial Intelligence
- Computer Science, Interdisciplinary Applications
- Computer Science, Theory & Methods