2024/08/09 by Yamin Sepehri, Pedram Pad, Sepehri, Yamin +5 · 1 citation
Computer Science · #Artificial neural network #Cloud computing #Deep learning #Deep neural networks #Differential privacy #Edge device #Enhanced Data Rates for GSM Evolution #Noise (video) #Privacy-Preserving Technologies in Data #Training (meteorology)
paper · pdf · doi:10.1109/tmm.2025.3607801
published in IEEE Transactions on Multimedia 27, 8803-8818 (Institute of Electrical and Electronics Engineers)
openalex publication_date 2025/01/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/06/11
The training phase of deep neural networks requires substantial resources and as such is often performed on cloud servers. However, this raises privacy concerns when the training dataset contains sensitive content, e.g., facial or medical images. In this work, we propose a method to perform the training phase of a deep learning model on both an edge device and a cloud server that prevents sensitive content being transmitted to the cloud while retaining the desired information. The proposed privacy-preserving method uses adversarial early exits to suppress the sensitive content at the edge and transmits the task-relevant information to the cloud. This approach incorporates noise addition during the training phase to provide a differential privacy guarantee. We extensively test our method on different facial and medical datasets with diverse attributes using various deep learning architectures, showcasing its outstanding performance. We also demonstrate the effectiveness of privacy preservation through successful defenses against different white-box, deep and GAN-based reconstruction attacks. This approach is designed for resource-constrained edge devices, ensuring minimal memory usage and computational overhead.