Privacy-Accuracy Trade-Off of Inference as Service
Privacy-Accuracy Trade-Off of Inference as Service
复制标题
DOI：
10.1109/icassp39728.2021.9413438
复制
发表时间：
2021-06
期刊：
影响因子：
--
通讯作者：
Yulu Jin;L. Lai
中科院分区：
文献类型：
--
作者：
Yulu Jin;L. Lai
在本文中，我们提出了一个通用的框架，以提供一个理想的权衡推理的准确性和隐私保护的推理服务的情况。用户将通过隐私保护映射来预处理数据，而不是直接将数据发送到服务器，这将增加隐私保护，但降低推理准确性。为了妥善解决隐私保护和推理准确性之间的权衡，我们制定了一个优化问题，以找到最佳的隐私保护映射。即使问题是非凸的一般情况下，我们的问题的特征结构，并开发一个迭代算法来找到所需的隐私保护映射。
In this paper, we propose a general framework to provide a desirable trade-off between inference accuracy and privacy protection in the inference as service scenario. Instead of sending data directly to the server, the user will preprocess the data through a privacy-preserving mapping, which will increase privacy protection but reduce inference accuracy. To properly address the trade-off between privacy protection and inference accuracy, we formulate an optimization problem to find the optimal privacy-preserving mapping. Even though the problem is non-convex in general, we characterize nice structures of the problem and develop an iterative algorithm to find the desired privacy-preserving mapping.