PINFER: Privacy-Preserving Inference for Machine Learning

10/04/2019 ∙ by Marc Joye, et al. ∙ 0 ∙ share

The foreseen growing role of outsourced machine learning services is raising concerns about the privacy of user data. Several technical solutions are being proposed to address the issue. Hardware security modules in cloud data centres appear limited to enterprise customers due to their complexity, while general multi-party computation techniques require a large number of message exchanges. This paper proposes a variety of protocols for privacy-preserving regression and classification that (i) only require additively homomorphic encryption algorithms, (ii) limit interactions to a mere request and response, and (iii) that can be used directly for important machine-learning algorithms such as logistic regression and SVM classification. The basic protocols are then extended and applied to feed-forward neural networks.

READ FULL TEXT

PrivLogit: Efficient Privacy-preserving Logistic Regression by Tailoring Numerical Optimizers

Safeguarding privacy in machine learning is highly desirable, especially...
11/03/2016 ∙ by Wei Xie, et al. ∙ 0 ∙ share

read it
Trident: Efficient 4PC Framework for Privacy Preserving Machine Learning

Machine learning has started to be deployed in fields such as healthcare...
12/05/2019 ∙ by Rahul Rachuri, et al. ∙ 0 ∙ share

read it
CryptoNN: Training Neural Networks over Encrypted Data

Emerging neural networks based machine learning techniques such as deep ...
04/15/2019 ∙ by Runhua Xu, et al. ∙ 0 ∙ share

read it
BLAZE: Blazing Fast Privacy-Preserving Machine Learning

Machine learning tools have illustrated their potential in many signific...
04/06/2020 ∙ by Ajith Suresh, et al. ∙ 0 ∙ share

read it
Reversible Privacy Preservation using Multi-level Encryption and Compressive Sensing

Security monitoring via ubiquitous cameras and their more extended in in...
06/20/2019 ∙ by Mehmet Yamac, et al. ∙ 0 ∙ share

read it
Privacy-Preserving Gaussian Process Regression – A Modular Approach to the Application of Homomorphic Encryption

Much of machine learning relies on the use of large amounts of data to t...
01/28/2020 ∙ by Peter Fenner, et al. ∙ 17 ∙ share

read it
Privacy Preserving K-Means Clustering: A Secure Multi-Party Computation Approach

Knowledge discovery is one of the main goals of Artificial Intelligence....
09/22/2020 ∙ by Daniel Hurtado Ramírez, et al. ∙ 0 ∙ share

read it

PINFER: Privacy-Preserving Inference for Machine Learning

Authors

PrivLogit: Efficient Privacy-preserving Logistic Regression by Tailoring Numerical Optimizers

Trident: Efficient 4PC Framework for Privacy Preserving Machine Learning

CryptoNN: Training Neural Networks over Encrypted Data

BLAZE: Blazing Fast Privacy-Preserving Machine Learning

Reversible Privacy Preservation using Multi-level Encryption and Compressive Sensing

Privacy-Preserving Gaussian Process Regression – A Modular Approach to the Application of Homomorphic Encryption

Privacy Preserving K-Means Clustering: A Secure Multi-Party Computation Approach

This week in AI