EUVIMEDEUVIMED entdecken
Crossref

Influence-Driven Data Poisoning for Robust Recommender Systems

Chenwang Wu, Defu Lian, Yong Ge, Zhihao Zhu, Enhong Chen

IEEE Transactions on Pattern Analysis and Machine Intelligence · 2023-10 · 45 (10) · 11915-11931

DOI: 10.1109/tpami.2023.3274759 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023-10. Verfasst von Chenwang Wu und 4 weiteren Autor:innen. Die Arbeit wurde bisher 47-mal zitiert (Crossref) und nennt 63 Literaturangaben. Gefördert von: National Key R&D Program of China, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recent studies have shown that recommender systems are vulnerable, and it is easy for attackers to inject well-designed malicious profiles into the system, resulting in biased recommendations. We cannot deprive these data's injection right and deny their existence's rationality, making it imperative to study recommendation robustness. Despite impressive emerging work, threat assessment of the bi-level poisoning problem and the imperceptibility of poisoning users remain key challenges to be solved. To this end, we propose Infmix, an efficient poisoning attack strategy. Specifically, Infmix consists of an influence-based threat estimator and a user generator, Usermix. First, the influence-based estimator can efficiently evaluate the user's harm to the recommender system without retraining, which is challenging for existing attacks. Second, Usermix, a distribution-agnostic generator, can generate unnoticeable fake data even with a few known users. Under the guidance of the threat estimator, Infmix can select the users with large attacking impacts from the quasi-real candidates generated by Usermix. Extensive experiments demonstrate Infmix's superiority by attacking six recommendation systems with four real datasets. Additionally, we propose a novel defense strategy, adversarial poisoning training (APT). It mimics the poisoning process by injecting fake users (ERM users) committed to minimizing empirical risk to build a robust system. Similar to Infmix, we also utilize the influence function to solve the bi-level optimization challenge of generating ERM users. Although the idea of "fighting fire with fire" in APT seems counterintuitive, we prove its effectiveness in improving recommendation robustness through theoretical analysis and empirical experiments.

Förderung

  • National Key R&D Program of China
  • National Natural Science Foundation of China

Zitieren

APA

Wu, C., Lian, D., Ge, Y., Zhu, Z., & Chen, E. (2023). Influence-Driven Data Poisoning for Robust Recommender Systems. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(10), 11915-11931. https://doi.org/10.1109/tpami.2023.3274759

Vancouver

Wu C, Lian D, Ge Y, Zhu Z, Chen E. Influence-Driven Data Poisoning for Robust Recommender Systems. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2023;45(10):11915-11931. doi:10.1109/tpami.2023.3274759

BibTeX
@article{Wu2023,
  author = {Wu, Chenwang and Lian, Defu and Ge, Yong and Zhu, Zhihao and Chen, Enhong},
  title = {Influence-Driven Data Poisoning for Robust Recommender Systems},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year = {2023},
  volume = {45},
  number = {10},
  pages = {11915--11931},
  doi = {10.1109/tpami.2023.3274759}
}
Kostenlos starten →