arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2105.08511cs.LGcs.CReess.IV
收藏

基于梯度对齐的隐私保护约束域泛化

Privacy-Preserving Constrained Domain Generalization via Gradient Alignment

Chris Xing Tian, Haoliang Li, Yufei Wang, Shiqi Wang

AI总结:

本文针对医学影像分类中数据稀缺与隐私限制导致的跨域泛化困难,提出一种隐私保护的约束域泛化方法,通过在服务器端引入梯度对齐损失改进信息聚合,并结合MMD理论解释其有效性,实验表明该方法优于现有联邦学习方法。

摘要:

深度神经网络(DNN)在医学影像应用中已展现出前所未有的成功。然而,由于数据集可用性有限以及患者隐私保护方面严格的法律和伦理要求,由大规模训练数据驱动的DNN在医学影像分类中的广泛应用在很大程度上受到了阻碍。例如,当仅使用一个域(如仅来自一家医院的数据)训练DNN时,其泛化到另一个域(如另一家医院的数据)的能力可能严重不足。在本文中,我们旨在通过开发一种隐私保护的约束域泛化方法来解决这一问题,目标是在隐私保护条件下提升泛化能力。具体而言,我们提出在中心服务器端利用一种新颖的梯度对齐损失来改进信息聚合过程,期望训练后的模型能够更好地泛化到“未见”但相关的医学影像。我们提出的方法的合理性和有效性可以通过将其与已被广泛用作分布距离度量的最大均值差异(MMD)建立联系来加以解释。在两个具有挑战性的医学影像分类任务上的实验结果表明,与最先进的联邦学习方法相比,我们的方法能够取得更好的跨域泛化能力。

英文摘要:

Deep neural networks (DNN) have demonstrated unprecedented success for medical imaging applications. However, due to the issue of limited dataset availability and the strict legal and ethical requirements for patient privacy protection, the broad applications of medical imaging classification driven by DNN with large-scale training data have been largely hindered. For example, when training the DNN from one domain (e.g., with data only from one hospital), the generalization capability to another domain (e.g., data from another hospital) could be largely lacking. In this paper, we aim to tackle this problem by developing the privacy-preserving constrained domain generalization method, aiming to improve the generalization capability under the privacy-preserving condition. In particular, We propose to improve the information aggregation process on the centralized server-side with a novel gradient alignment loss, expecting that the trained model can be better generalized to the "unseen" but related medical images. The rationale and effectiveness of our proposed method can be explained by connecting our proposed method with the Maximum Mean Discrepancy (MMD) which has been widely adopted as the distribution distance measurement. Experimental results on two challenging medical imaging classification tasks indicate that our method can achieve better cross-domain generalization capability compared to the state-of-the-art federated learning methods.

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。

↑