Poisoning Attacks on Federated Learning for Autonomous Driving

arXiv:2405.01073v1 Announce Type: new
Abstract: Federated Learning (FL) is a decentralized learning paradigm, enabling parties to collaboratively train models while keeping their data confidential. Within autonomous driving, it brings the potential of reducing data storage costs, reducing bandwidth requirements, and to accelerate the learning. FL is, however, susceptible to poisoning attacks. In this paper, we introduce two novel poisoning attacks on FL tailored to regression tasks within autonomous driving: FLStealth and Off-Track Attack (OTA). FLStealth, an untargeted attack, aims at …

abstract arxiv attacks autonomous autonomous driving bandwidth costs cs.ai cs.cr cs.cv cs.lg data data storage decentralized driving enabling federated learning however paradigm parties poisoning attacks requirements storage storage costs train type while