Precise Extraction of Deep Learning Models via Side-Channel Attacks on Edge/Endpoint Devices

arXiv:2403.02870v1 Announce Type: cross
Abstract: With growing popularity, deep learning (DL) models are becoming larger-scale, and only the companies with vast training datasets and immense computing power can manage their business serving such large models. Most of those DL models are proprietary to the companies who thus strive to keep their private models safe from the model extraction attack (MEA), whose aim is to steal the model by training surrogate models. Nowadays, companies are inclined to offload the models from …

arxiv attacks business can channel companies computing cs.ai cs.cr cs.lg datasets deep learning devices edge endpoint endpoint devices extraction large manage power scale side-channel side-channel attacks training vast