Towards Privacy-Preserving Split Learning: Destabilizing Adversarial Inference and Reconstruction Attacks in the Cloud

2025/02/28 by Higgins, Griffin, Razavi-Far, Roozbeh, Zhang, Xichen +3 · 1 citation
#Cryptography and Security (cs.CR) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2502.20629

Abstract

This work aims to provide both privacy and utility within a split learning framework while considering both forward attribute inference and backward reconstruction attacks. To address this, a novel approach has been proposed, which makes use of class activation maps and autoencoders as a plug-in strategy aiming to increase the user's privacy and destabilize an adversary. The proposed approach is compared with a dimensionality-reduction-based plug-in strategy, which makes use of principal component analysis to transform the feature map onto a lower-dimensional feature space. Our work shows that our proposed autoencoder-based approach is preferred as it can provide protection at an earlier split position over the tested architectures in our setting, and, hence, better utility for resource-constrained devices in edge-cloud collaborative inference (EC) systems.

Cited by

• Verifiable Split Learning via zk-SNARKs

Related

• Make Split, not Hijack: Preventing Feature-Space Hijacking Attacks in Split Learning
• Protecting Split Learning by Potential Energy Loss
• Evaluating Privacy Leakage in Split Learning
• A Stealthy Wrongdoer: Feature-Oriented Reconstruction Attack against Split Learning
• A More Secure Split: Enhancing the Security of Privacy-Preserving Split Learning