Make Split, not Hijack: Preventing Feature-Space Hijacking Attacks in Split Learning

arXiv:2404.09265v1 Announce Type: new
Abstract: The popularity of Machine Learning (ML) makes the privacy of sensitive data more imperative than ever. Collaborative learning techniques like Split Learning (SL) aim to protect client data while enhancing ML processes. Though promising, SL has been proved to be vulnerable to a plethora of attacks, thus raising concerns about its effectiveness on data privacy. In this work, we introduce a hybrid approach combining SL and Function Secret Sharing (FSS) to ensure client data privacy. …

aim arxiv attacks client cs.ai cs.cr data feature hijack hijacking machine machine learning privacy processes protect sensitive sensitive data space split learning techniques vulnerable