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.
@article{arxiv.2502.20629,
title = {Towards Privacy-Preserving Split Learning: Destabilizing Adversarial Inference and Reconstruction Attacks in the Cloud},
author = {Griffin Higgins and Roozbeh Razavi-Far and Xichen Zhang and Amir David and Ali Ghorbani and Tongyu Ge},
journal= {arXiv preprint arXiv:2502.20629},
year = {2025}
}