Poisoning Federated Recommender Systems with Fake Users

arXiv:2402.11637v1 Announce Type: new
Abstract: Federated recommendation is a prominent use case within federated learning, yet it remains susceptible to various attacks, from user to server-side vulnerabilities. Poisoning attacks are particularly notable among user-side attacks, as participants upload malicious model updates to deceive the global model, often intending to promote or demote specific targeted items. This study investigates strategies for executing promotion attacks in federated recommender systems.
Current poisoning attacks on federated recommender systems often rely on additional information, such …

arxiv attacks case cs.cr cs.ir cs.lg fake federated federated learning global malicious poisoning poisoning attacks recommender systems server systems updates upload use case vulnerabilities