Principled Approaches for Private Adaptation from a Public Source
Principled Approaches for Private Adaptation from a Public Source
复制标题
DOI：
--
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
Raef Bassily;M. Mohri;A. Suresh
中科院分区：
文献类型：
--
作者：
Raef Bassily;M. Mohri;A. Suresh
各种应用中的一个关键问题是从公共源域到私有目标域的域适配，对于公共源域，可以使用相对大量的没有隐私约束的标记数据，对于私有目标域，可以获得具有很少或没有标记数据的私有样本。在源或目标数据没有隐私约束的回归问题中，差异最小化方法的性能优于其他一些自适应算法基线。在这种方法的基础上，我们启动了从具有公共标记数据的源域到具有未标记私有数据的目标域的差异私有适配的原则性研究。针对这个问题，我们设计了基于差分私有差异的自适应算法。我们私人算法的设计和分析关键取决于我们证明的加权偏差的光滑逼近的几个关键性质，例如它相对于(CID：96)1范数的光滑性和它的梯度的敏感性。我们形式化地证明了我们的自适应算法得益于强的泛化和隐私保证。fi。
A key problem in a variety of applications is that of domain adaptation from a public source domain, for which a relatively large amount of labeled data with no privacy constraints is at one’s disposal, to a private target domain, for which a private sample is available with very few or no labeled data. In regression problems, where there are no privacy constraints on the source or target data, a discrepancy minimization approach was shown to outperform a number of other adaptation algorithm baselines. Building on that approach, we initiate a principled study of differentially private adaptation from a source domain with public labeled data to a target domain with unlabeled private data. We design differentially private discrepancy-based adaptation algorithms for this problem. The design and analysis of our private algorithms critically hinge upon several key properties we prove for a smooth approximation of the weighted discrepancy, such as its smoothness with respect to the (cid:96) 1 -norm and the sensitivity of its gradient. We formally show that our adaptation algorithms benefit from strong generalization and privacy guarantees.