EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

GADD: Game-Inspired Adversarial Distillation for Robust Graph Defense

Yabin Peng, Chenyu Zhou, Yuchen Liu, Kunlin Li, Fan Zhang, Shaoxun Liu

Information · 26.05.2026 · 17 (6) · 527

DOI: 10.3390/info17060527 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Information, online erschienen am 26.05.2026. Verfasst von Yabin Peng und 5 weiteren Autor:innen. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Jiangsu Province Science and Technology Department, Southeast University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Graph neural networks (GNNs) are highly effective on relational data, yet their performance degrades sharply when graph topology is poisoned before training. Existing defenses usually assume a fixed attack pattern and a fixed graph structure, which makes them brittle when the poisoned graph changes across attacks, perturbation budgets, or deployment conditions. We propose GADD, a game-inspired adversarial distillation framework for robust graph defense. GADD first constructs multiple positive and negative graph views through a homophily-aware graph sampling scheme, allowing the model to learn from both purified and high-risk subgraphs. It then trains a heterogeneous group of student GNNs online, where each student receives global class-distribution knowledge from its peers and local structural knowledge through an adversarial cyclic distillation objective. Finally, GADD replaces uniform ensembling with an entropy-regularized adaptive aggregation rule that assigns graph-adaptive weights according to confidence and inter-model agreement. On Cora, CiteSeer, and PubMed, GADD consistently improves robustness against both Meta and Nettack attacks while preserving clean accuracy. Under the strongest Meta and Nettack settings in the main benchmark, GADD improves the best competing baseline by up to 2.99 and 3.42 percentage points, respectively. Additional ablations show that graph sampling, adversarial distillation, and adaptive aggregation all contribute materially to the final robustness gains.

Förderung

  • Jiangsu Province Science and Technology Department
  • Southeast University

Zitieren

APA

Peng, Y., Zhou, C., Liu, Y., Li, K., Zhang, F., & Liu, S. (2026). GADD: Game-Inspired Adversarial Distillation for Robust Graph Defense. Information, 17(6), 527. https://doi.org/10.3390/info17060527

Vancouver

Peng Y, Zhou C, Liu Y, Li K, Zhang F, Liu S. GADD: Game-Inspired Adversarial Distillation for Robust Graph Defense. Information. 2026;17(6):527. doi:10.3390/info17060527

BibTeX
@article{Peng2026,
  author = {Peng, Yabin and Zhou, Chenyu and Liu, Yuchen and Li, Kunlin and Zhang, Fan and Liu, Shaoxun},
  title = {GADD: Game-Inspired Adversarial Distillation for Robust Graph Defense},
  journal = {Information},
  year = {2026},
  volume = {17},
  number = {6},
  pages = {527},
  doi = {10.3390/info17060527}
}
Kostenlos starten →