EUVIMEDEUVIMED entdecken
Crossref

On Stealing Graph Neural Network Models

Marcin Podhajski, Jan Dubiński, Franziska Boenisch, Adam Dziedzic, Agnieszka Pręgowska, Tomasz Paweł Michalak

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (29) · 24846-24854

DOI: 10.1609/aaai.v40i29.39671 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Marcin Podhajski und 5 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Current graph neural network (GNN) model-stealing methods rely heavily on queries to the victim model, assuming no hard query limits. However, in reality, the number of allowed queries can be severely limited. In this paper, we demonstrate how an adversary can extract a GNN with very limited interactions with the model. Our approach first enables the adversary to obtain the model backbone without making direct queries to the victim model and then to strategically utilize a fixed query limit to extract the most informative data. The experiments on eight real-world datasets demonstrate the effectiveness of the attack, even under a very restricted query limit and under defense against model extraction in place. Our findings underscore the need for robust defenses against GNN model extraction threats.

Zitieren

APA

Podhajski, M., Dubiński, J., Boenisch, F., Dziedzic, A., Pręgowska, A., & Michalak, T. P. (2026). On Stealing Graph Neural Network Models. Proceedings of the AAAI Conference on Artificial Intelligence, 40(29), 24846-24854. https://doi.org/10.1609/aaai.v40i29.39671

Vancouver

Podhajski M, Dubiński J, Boenisch F, Dziedzic A, Pręgowska A, Michalak TP. On Stealing Graph Neural Network Models. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(29):24846-24854. doi:10.1609/aaai.v40i29.39671

BibTeX
@article{Podhajski2026,
  author = {Podhajski, Marcin and Dubiński, Jan and Boenisch, Franziska and Dziedzic, Adam and Pręgowska, Agnieszka and Michalak, Tomasz Paweł},
  title = {On Stealing Graph Neural Network Models},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {29},
  pages = {24846--24854},
  doi = {10.1609/aaai.v40i29.39671}
}
Kostenlos starten →