We introduce PrivPy, a practical privacy-preserving collaborative computation
framework, especially optimized for machine learning tasks. PrivPy provides an
easy-to- use and highly compatible Python programming front- end which supports
high-level array operations and different secure computation engines to allow
for security assumptions and performance trade-offs. With PrivPy, programmers
can write modern machine learning algorithms conveniently and efficiently in
Python. We also design and implement a new efficient computation engine, with
which people can use competing cloud providers to efficiently perform general
arithmetics over real numbers. We demonstrate the usability and scalability of
PrivPy using common machine learning models (e.g. logistic regression and
convolutional neural networks) and real- world datasets (including a
5000-by-1-million matrix).
Description
PrivPy: Enabling Scalable and General Privacy-Preserving Machine Learning
%0 Generic
%1 li2018privpy
%A Li, Yi
%A Duan, Yitao
%A Yu, Yu
%A Zhao, Shuoyao
%A Xu, Wei
%D 2018
%K PrivPy Privacy Security deepscan
%T PrivPy: Enabling Scalable and General Privacy-Preserving Machine
Learning
%U http://arxiv.org/abs/1801.10117
%X We introduce PrivPy, a practical privacy-preserving collaborative computation
framework, especially optimized for machine learning tasks. PrivPy provides an
easy-to- use and highly compatible Python programming front- end which supports
high-level array operations and different secure computation engines to allow
for security assumptions and performance trade-offs. With PrivPy, programmers
can write modern machine learning algorithms conveniently and efficiently in
Python. We also design and implement a new efficient computation engine, with
which people can use competing cloud providers to efficiently perform general
arithmetics over real numbers. We demonstrate the usability and scalability of
PrivPy using common machine learning models (e.g. logistic regression and
convolutional neural networks) and real- world datasets (including a
5000-by-1-million matrix).
@misc{li2018privpy,
abstract = {We introduce PrivPy, a practical privacy-preserving collaborative computation
framework, especially optimized for machine learning tasks. PrivPy provides an
easy-to- use and highly compatible Python programming front- end which supports
high-level array operations and different secure computation engines to allow
for security assumptions and performance trade-offs. With PrivPy, programmers
can write modern machine learning algorithms conveniently and efficiently in
Python. We also design and implement a new efficient computation engine, with
which people can use competing cloud providers to efficiently perform general
arithmetics over real numbers. We demonstrate the usability and scalability of
PrivPy using common machine learning models (e.g. logistic regression and
convolutional neural networks) and real- world datasets (including a
5000-by-1-million matrix).},
added-at = {2019-05-15T14:51:38.000+0200},
author = {Li, Yi and Duan, Yitao and Yu, Yu and Zhao, Shuoyao and Xu, Wei},
biburl = {https://www.bibsonomy.org/bibtex/286a72aa424a2967d731ec57a8a2b82ca/mittim},
description = {PrivPy: Enabling Scalable and General Privacy-Preserving Machine Learning},
interhash = {443f9e2989d5056ca4fc109f38e32579},
intrahash = {86a72aa424a2967d731ec57a8a2b82ca},
keywords = {PrivPy Privacy Security deepscan},
note = {cite arxiv:1801.10117},
timestamp = {2019-05-15T14:51:38.000+0200},
title = {PrivPy: Enabling Scalable and General Privacy-Preserving Machine
Learning},
url = {http://arxiv.org/abs/1801.10117},
year = 2018
}