Scalable Privacy-Preserving Distributed Learning

by   David Froelicher, et al.

In this paper, we address the problem of privacy-preserving distributed learning and evaluation of machine learning models by analyzing it in the widespread MapReduce abstraction that we extend with privacy constraints. Following this abstraction, we instantiate SPINDLE (Scalable Privacy-preservINg Distributed LEarning), an operational distributed system that supports the privacy-preserving training and evaluation of generalized linear models on distributed datasets. SPINDLE enables the efficient execution of distributed gradient descent while ensuring data and model confidentiality, as long as at least one of the data providers is honest-but-curious. The trained model is then used for oblivious predictions on confidential data. SPINDLE is able to efficiently perform demanding training tasks that require a high number of iterations on large input data with thousands of features, distributed among hundreds of data providers. It relies on a multiparty homomorphic encryption scheme to execute high-depth computations on encrypted data without significant overhead. It further leverages on its distributed construction and the packing capabilities of the cryptographic scheme to efficiently parallelize the computations at multiple levels. In our evaluation, SPINDLE performs the training of a logistic-regression model on a dataset of one million samples with 32 features distributed among 160 data providers in less than 176 seconds, yielding similar accuracy to non-secure centralized models.


page 9

page 13

page 14


POSEIDON: Privacy-Preserving Federated Neural Network Learning

In this paper, we address the problem of privacy-preserving training and...

Drynx: Decentralized, Secure, Verifiable System for Statistical Queries and Machine Learning on Distributed Datasets

Data sharing has become of primary importance in many domains such as bi...

Privacy Preserving Analytics on Distributed Medical Data

Objective: To enable privacy-preserving learning of high quality generat...

Scalable privacy-preserving cancer type prediction with homomorphic encryption

Machine Learning (ML) alleviates the challenges of high-dimensional data...

Privacy-Preserving Distributed Machine Learning Made Faster

With the development of machine learning, it is difficult for a single s...

Octopus: Privacy-Preserving Collaborative Evaluation of Loan Stacking

With the rise of online lenders, the loan stacking problem has become a ...

Secure and Distributed Assessment of Privacy-Preserving Releases of GWAS

Genome-wide association studies (GWAS) identify correlations between the...

Please sign up or login with your details

Forgot password? Click here to reset