Certifiable robustness and robust training for graph convolutional networks

Daniel Zügner, Stephan Günnemann

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

125 Zitate (Scopus)

Abstract

Recent works show that Graph Neural Networks (GNNs) are highly non-robust with respect to adversarial attacks on both the graph structure and the node attributes, making their outcomes unreliable. We propose the first method for certifiable (non-)robustness of graph convolutional networks with respect to perturbations of the node attributes1. We consider the case of binary node attributes (e.g. bag-of-words) and perturbations that are L0-bounded. If a node has been certified with our method, it is guaranteed to be robust under any possible perturbation given the attack model. Likewise, we can certify non-robustness. Finally, we propose a robust semi-supervised training procedure that treats the labeled and unlabeled nodes jointly. As shown in our experimental evaluation, our method significantly improves the robustness of the GNN with only minimal effect on the predictive accuracy.

OriginalspracheEnglisch
TitelKDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Herausgeber (Verlag)Association for Computing Machinery
Seiten246-256
Seitenumfang11
ISBN (elektronisch)9781450362016
DOIs
PublikationsstatusVeröffentlicht - 25 Juli 2019
Veranstaltung25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2019 - Anchorage, USA/Vereinigte Staaten
Dauer: 4 Aug. 20198 Aug. 2019

Publikationsreihe

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

Konferenz

Konferenz25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2019
Land/GebietUSA/Vereinigte Staaten
OrtAnchorage
Zeitraum4/08/198/08/19

Fingerprint

Untersuchen Sie die Forschungsthemen von „Certifiable robustness and robust training for graph convolutional networks“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren