EUVIMEDEUVIMED entdecken
Crossref

Multi-Agent Attacks for Black-Box Social Recommendations

Shijie Wang, Wenqi Fan, Xiao-Yong Wei, Xiaowei Mei, Shanru Lin, Qing Li

ACM Transactions on Information Systems · 31.01.2025 · 43 (1) · 1-26

DOI: 10.1145/3696105 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Information Systems, online erschienen am 19.12.2024. Verfasst von Shijie Wang und 5 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 64 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Hong Kong Research Grants Council, The Hong Kong Polytechnic University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rise of online social networks has facilitated the evolution of social recommender systems, which incorporate social relations to enhance users’ decision-making process. With the great success of Graph Neural Networks (GNNs) in learning node representations, GNN-based social recommendations have been widely studied to model user-item interactions and user-user social relations simultaneously. Despite their great successes, recent studies have shown that these advanced recommender systems are highly vulnerable to adversarial attacks, in which attackers can inject well-designed fake user profiles to disrupt recommendation performances. While most existing studies mainly focus on targeted attacks to promote target items on vanilla recommender systems, untargeted attacks to degrade the overall prediction performance are less explored on social recommendations under a black-box scenario. To perform untargeted attacks on social recommender systems, attackers can construct malicious social relationships for fake users to enhance the attack performance. However, the coordination of social relations and item profiles is challenging for attacking black-box social recommendations. To address this limitation, we first conduct several preliminary studies to demonstrate the effectiveness of cross-community connections and cold-start items in degrading recommendations performance. Specifically, we propose a novel framework MultiAttack based on multi-agent reinforcement learning to coordinate the generation of cold-start item profiles and cross-community social relations for conducting untargeted attacks on black-box social recommendations. Comprehensive experiments on various real-world datasets demonstrate the effectiveness of our proposed attacking framework under the black-box setting.

Förderung

  • National Natural Science Foundation of China
  • Hong Kong Research Grants Council
  • The Hong Kong Polytechnic University
  • SHTM Interdisciplinary Large Grant

Zitieren

APA

Wang, S., Fan, W., Wei, X., Mei, X., Lin, S., & Li, Q. (2025). Multi-Agent Attacks for Black-Box Social Recommendations. ACM Transactions on Information Systems, 43(1), 1-26. https://doi.org/10.1145/3696105

Vancouver

Wang S, Fan W, Wei X, Mei X, Lin S, Li Q. Multi-Agent Attacks for Black-Box Social Recommendations. ACM Transactions on Information Systems. 2025;43(1):1-26. doi:10.1145/3696105

BibTeX
@article{Wang2025,
  author = {Wang, Shijie and Fan, Wenqi and Wei, Xiao-Yong and Mei, Xiaowei and Lin, Shanru and Li, Qing},
  title = {Multi-Agent Attacks for Black-Box Social Recommendations},
  journal = {ACM Transactions on Information Systems},
  year = {2025},
  volume = {43},
  number = {1},
  pages = {1--26},
  doi = {10.1145/3696105}
}
Kostenlos starten →