通信学报2025，Vol.46Issue(3)：174-186,13.DOI:10.11959/j.issn.1000-436x.2025056
基于代理生成对抗网络的服务质量感知云API推荐系统投毒攻击
Poisoning attack on quality of service aware cloud API recommender system via surrogate generative adversarial network
摘要
Abstract
To address the shortcomings of existing poisoning attack methods,where the generated fake user attack data suf-fers from poor attack effectiveness and high detectability,a poisoning attack method based on the surrogate generative ad-versarial network(S-GAN)was proposed.Firstly,K-means was used in the generative adversarial network to classify the data,and a self-attention mechanism was incorporated to learn the global features within each class,solving the problem of difficulty in capturing and mimicking key features of real users in sparse data,thereby enhancing the concealment of fake users.Secondly,a surrogate model was deployed to evaluate the attack effectiveness of the GAN-generated fake users,and the evaluation results were employed as a surrogate loss to optimize the GAN,thereby facilitating the attack effectiveness while considering the concealment of fake users.Experiments conducted on cloud API quality of service datasets demon-strate that the proposed method outperforms existing methods in balancing the effectiveness and concealment of attacks.关键词
推荐系统/云API/投毒攻击/生成对抗网络/代理模型Key words
recommender system/cloud API/poisoning attack/generative adversarial network/surrogate model分类
电子信息工程引用本文复制引用
陈真,刘伟,吕瑞民,马佳洁,冯佳音,尤殿龙..基于代理生成对抗网络的服务质量感知云API推荐系统投毒攻击[J].通信学报,2025,46(3):174-186,13.基金项目
国家自然科学基金资助项目(No.62102348,No.62276226) （No.62102348,No.62276226）
河北省自然科学基金资助项目(No.F2022203012) （No.F2022203012）
河北省科技计划基金资助项目(No.236Z0103G,No.236Z7725G) （No.236Z0103G,No.236Z7725G）
河北省创新能力提升计划基金资助项目(No.22567626H) The National Natural Science Foundation of China(No.62102348,No.62276226),The Natural Science Foundation of Hebei Province(No.F2022203012),The Science and Technology Program of Hebei Province(No.236Z0103G,No.236Z7725G),The Innovation Capability Improvement Plan Project of Hebei Province(No.22567626H) （No.22567626H）
