arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.01502stat.MLcs.CRcs.LG
收藏

Sage差异私有ML平台中的隐私计费和质量控制

Privacy Accounting and Quality Control in the Sage Differentially Private ML Platform

Columbia University(哥伦比亚大学)

Mathias Lecuyer, Riley Spahn, Kiran Vodrahalli, Roxana Geambasu, Daniel Hsu

摘要:

公司越来越多地将在敏感用户数据上训练的机器学习(ML)模型暴露给不受信任的域,例如最终用户设备和广泛访问的模型存储。我们介绍了Sage,一个差分私有(DP)ML平台,它通过模型限制了训练数据的累积泄漏。Sage基于DP ML算法的丰富文献,为全球DP的两个最紧迫的系统挑战提供实用的解决方案:隐私预算耗尽和隐私-效用权衡。为了解决前者,我们开发了块组合,这是一种新的隐私损失核算方法,它利用ML工作负载不断增长的数据库机制,在敏感数据流上无休止地保持训练模型,同时为该数据流强制执行全局DP保证。为了解决后者,我们开发了隐私自适应训练,这是一个在不断增长的数据量和/或不断增加的隐私参数上训练模型的过程,直到模型很有可能满足开发人员配置的质量标准。它们说明了系统如何专注于ML工作负载的特征,从而实现务实的解决方案,而当人们专注于单个算法时,这些解决方案并不明显,正如大多数DP ML文献所做的那样。

英文摘要:

Companies increasingly expose machine learning (ML) models trained over sensitive user data to untrusted domains, such as end-user devices and wide-access model stores. We present Sage, a differentially private (DP) ML platform that bounds the cumulative leakage of training data through models. Sage builds upon the rich literature on DP ML algorithms and contributes pragmatic solutions to two of the most pressing systems challenges of global DP: running out of privacy budget and the privacy-utility tradeoff. To address the former, we develop block composition, a new privacy loss accounting method that leverages the growing database regime of ML workloads to keep training models endlessly on a sensitive data stream while enforcing a global DP guarantee for the stream. To address the latter, we develop privacy-adaptive training, a process that trains a model on growing amounts of data and/or with increasing privacy parameters until, with high probability, the model meets developer-configured quality criteria. They illustrate how a systems focus on characteristics of ML workloads enables pragmatic solutions that are not apparent when one focuses on individual algorithms, as most DP ML literature does.

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。

↑