Certifiable Robustness of Graph Convolutional Networks under Structure Perturbations

Daniel Zügner, Stephan Günnemann

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

46 Zitate (Scopus)

Abstract

Recent works show that message-passing neural networks (MPNNs) can be fooled by adversarial attacks on both the node attributes and the graph structure. Since MPNNs are currently being rapidly adopted in real-world applications, it is thus crucial to improve their reliablility and robustness. While there has been progress on robustness certification of MPNNs under perturbation of the node attributes, no existing method can handle structural perturbations. These perturbations are especially challenging because they alter the message passing scheme itself. In this work we close this gap and propose the first method to certify robustness of Graph Convolutional Networks (GCNs) under perturbations of the graph structure. We show how this problem can be expressed as a jointly constrained bilinear program - a challenging, yet well-studied class of problems - and propose a novel branch-and-bound algorithm to obtain lower bounds on the global optimum. These lower bounds are significantly tighter and can certify up to twice as many nodes compared to a standard linear relaxation.

OriginalspracheEnglisch
TitelKDD 2020 - Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Herausgeber (Verlag)Association for Computing Machinery
Seiten1656-1665
Seitenumfang10
ISBN (elektronisch)9781450379984
DOIs
PublikationsstatusVeröffentlicht - 23 Aug. 2020
Veranstaltung26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2020 - Virtual, Online, USA/Vereinigte Staaten
Dauer: 23 Aug. 202027 Aug. 2020

Publikationsreihe

NameProceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Konferenz

Konferenz26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2020
Land/GebietUSA/Vereinigte Staaten
OrtVirtual, Online
Zeitraum23/08/2027/08/20

Fingerprint

Untersuchen Sie die Forschungsthemen von „Certifiable Robustness of Graph Convolutional Networks under Structure Perturbations“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren