arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.05092cs.CVcs.CRcs.DCcs.LGeess.IV
收藏

PriPHiT: 保护隐私的深度神经网络分层训练

PriPHiT: Privacy-Preserving Hierarchical Training of Deep Neural Networks

Signal Processing Laboratory (LTS4), École Polytechnique Fédérale de Lausanne (EPFL)(瑞士联邦理工学院洛桑校区信号处理实验室);Centre Suisse d’Electronique et de Microtechnique (CSEM)(瑞士电子与微技术中心)

Yamin Sepehri, Pedram Pad, Pascal Frossard, L. Andrea Dunbar

AI总结:

PriPHiT通过在边缘设备和云服务器上分层训练深度神经网络,利用对抗性早期退出和噪声添加技术,在保护隐私的同时提升模型性能。

摘要:

深度神经网络的训练阶段需要大量资源,因此通常在云服务器上进行。然而,当训练数据集包含敏感内容(例如面部或医疗图像)时,这会引发隐私问题。在本文中,我们提出了一种方法,在边缘设备和云服务器上进行深度学习模型的训练,以防止敏感内容传输到云中,同时保留所需信息。所提出的隐私保护方法利用对抗性早期退出来在边缘端抑制敏感内容,并将任务相关的信息传输到云中。该方法在训练阶段加入噪声以提供差分隐私保障。我们广泛测试了该方法在不同面部和医疗数据集上的性能,使用各种深度学习架构,展示了其卓越的性能。我们还通过成功防御不同白盒、深度和GAN基的重建攻击,证明了隐私保护的有效性。该方法适用于资源受限的边缘设备,确保最小的内存使用和计算开销。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑