Distributed and Secure ML with Self-tallying Multi-party Aggregation

TL;DR

This paper introduces a distributed, privacy-preserving machine learning framework that enables untrusted users to collaboratively train models securely without private channels or trusted third parties, using homomorphic addition and zero-knowledge proofs.

Contribution

The work presents a novel framework combining homomorphic addition and zero-knowledge proofs for secure, distributed ML that is robust against malicious data contributions and does not require trusted third parties.

Findings

Framework supports various ML algorithms like LDA, Naive Bayes, and Decision Trees.

Ensures data privacy and integrity in untrusted environments.

Eliminates need for private channels and trusted third parties.

Abstract

Privacy preserving multi-party computation has many applications in areas such as medicine and online advertisements. In this work, we propose a framework for distributed, secure machine learning among untrusted individuals. The framework consists of two parts: a two-step training protocol based on homomorphic addition and a zero knowledge proof for data validity. By combining these two techniques, our framework provides privacy of per-user data, prevents against a malicious user contributing corrupted data to the shared pool, enables each user to self-compute the results of the algorithm without relying on external trusted third parties, and requires no private channels between groups of users. We show how different ML algorithms such as Latent Dirichlet Allocation, Naive Bayes, Decision Trees etc. fit our framework for distributed, secure computing.