AGAS

推薦システムに対する効率的かつ効果的なエージェント型グループ・シリング攻撃 これは論文「An Efficient and Effective Agentic Group Shilling Attack on Recommender Systems」の公式コードです。本論文ではAGASを紹介します。これは、ブラックボックスの協調フィルタリング推薦システムに対するLLM駆動のシリング攻撃です。1つの Coordinator が、一連のラウンドにわたって偽ユーザーのworker プールを統制します。各ラウンドで、Coordinatorは8つの戦略のうち1つを選択し、...

Phantom Subgroup Poisoning: Stealth Attacks on Federated Recommender Systems

Federated recommender systems FedRec have emerged as a promising solution for delivering personalized recommendations while safeguarding user privacy. However, recent studies have demonstrated their vulnerability to poisoning attacks. Existing attacks typically target the entire user group, which...

RAID: an In-Training Defense against Attribute Inference Attacks in Recommender Systems

In various networks and mobile applications, users are highly susceptible to attribute inference attacks, with particularly prevalent occurrences in recommender systems. Attackers exploit partially exposed user profiles in recommendation models, such as user embeddings, to infer private attribute...