PrivPy: General and Scalable Privacy-Preserving Data Mining

被引：15

作者：

Li, Yi ^{[1
]}

Xu, Wei ^{[1
]}

机构：

[1] Tsinghua Univ, Beijing, Peoples R China

来源：

KDD'19: PROCEEDINGS OF THE 25TH ACM SIGKDD INTERNATIONAL CONFERENCCE ON KNOWLEDGE DISCOVERY AND DATA MINING | 2019年

关键词：

privacy-preserving; data mining; numpy; !text type='python']python[!/text;

D O I：

10.1145/3292500.3330920

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Privacy is a big hurdle for collaborative data mining across multiple parties. We present multi-party computation (MPC) framework designed for large-scale data mining tasks. PrivPy combines an easy-to-use and highly flexible Python programming interface with state-of-the-art secret-sharing-based MPC backend. With essential data types and operations (such as NumPy arrays and broadcasting), as well as automatic code-rewriting, programmers can write modern data mining algorithms conveniently in familiar Python. We demonstrate that we can support many real-world machine learning algorithms (e.g. logistic regression and convolutional neural networks) and large datasets (e.g. 5000-by-1-million matrix) with minimal algorithm porting effort.

引用

页码：1299 / 1307

页数：9