arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.01022cs.CRcs.IR
收藏

推荐系统中的投毒攻击与防御:综述

Poisoning Attacks and Defenses in Recommender Systems: A Survey

Zongwei Wang, Junliang Yu, Min Gao, Wei Yuan, Guanhua Ye, Shazia Sadiq, Hongzhi Yin

AI总结:

本综述从攻击者视角系统梳理了推荐系统中的投毒攻击流程(含设定目标、评估能力、分析架构与实施策略),并将防御策略分为数据过滤与鲁棒训练,指出了现有局限与未来方向。

摘要:

现代推荐系统(RS)极大地提升了数字平台上的用户体验,但面临着来自投毒攻击的严重威胁。这些攻击旨在通过不道德的手段操纵推荐输出,通过注入恶意数据或干预模型训练来利用RS中的漏洞。本综述从攻击者的视角审视这些威胁,为其机制和影响提供了新的见解。具体而言,我们详细阐述了一个系统化流程,包含投毒攻击的四个阶段:设定攻击目标、评估攻击者能力、分析受害者架构以及实施投毒策略。该流程不仅与各种攻击战术相契合,还作为一个全面的分类法,以准确定位不同投毒攻击的侧重点。相应地,我们从防御者视角进一步将防御策略分为两大类:投毒数据过滤和鲁棒训练。最后,我们强调了现有的局限性,并为该领域的进一步探索提出了创新方向。

英文摘要:

Modern recommender systems (RS) have profoundly enhanced user experience across digital platforms, yet they face significant threats from poisoning attacks. These attacks, aimed at manipulating recommendation outputs for unethical gains, exploit vulnerabilities in RS through injecting malicious data or intervening model training. This survey presents a unique perspective by examining these threats through the lens of an attacker, offering fresh insights into their mechanics and impacts. Concretely, we detail a systematic pipeline that encompasses four stages of a poisoning attack: setting attack goals, assessing attacker capabilities, analyzing victim architecture, and implementing poisoning strategies. The pipeline not only aligns with various attack tactics but also serves as a comprehensive taxonomy to pinpoint focuses of distinct poisoning attacks. Correspondingly, we further classify defensive strategies into two main categories: poisoning data filtering and robust training from the defender's perspective. Finally, we highlight existing limitations and suggest innovative directions for further exploration in this field.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑