Abstract:While machine learning has become pervasive in as diversified fields as industry, healthcare, social networks, privacy concerns regarding the training data have gained a critical importance. In settings where several parties wish to collaboratively train a common model without jeopardizing their sensitive data, the need for a private training protocol is particularly stringent and implies to protect the data against both the model's end-users and the actors of the training phase. Differential privacy (DP) and cryptographic primitives are complementary popular countermeasures against privacy attacks. Among these cryptographic primitives, fully homomorphic encryption (FHE) offers ciphertext malleability at the cost of time-consuming operations in the homomorphic domain. In this paper, we design SHIELD, a probabilistic approximation algorithm for the argmax operator which is both fast when homomorphically executed and whose inaccuracy is used as a feature to ensure DP guarantees. Even if SHIELD could have other applications, we here focus on one setting and seamlessly integrate it in the SPEED collaborative training framework from "SPEED: Secure, PrivatE, and Efficient Deep learning" (Grivet Sébert et al., 2021) to improve its computational efficiency. After thoroughly describing the FHE implementation of our algorithm and its DP analysis, we present experimental results. To the best of our knowledge, it is the first work in which relaxing the accuracy of an homomorphic calculation is constructively usable as a degree of freedom to achieve better FHE performances.

SHAPER: A General Architecture for Privacy-Preserving Primitives in Secure Machine Learning.

Towards Secure and Practical Machine Learning Via Secret Sharing and Random Permutation

Poster: Nebula: an Industrial-purpose Privacy-preserving Machine Learning System

PPMLAC: High Performance Chipset Architecture for Secure Multi-Party Computation

PPMLAC

Cryptographic Approaches for Privacy-Preserving Machine Learning

A privacy preserving federated learning scheme using homomorphic encryption and secret sharing

Privacy-Preserving Machine Learning in Cloud-Edge-End Collaborative Environments

Efficient Privacy-Preserving Machine Learning with Lightweight Trusted Hardware

PrivColl: Practical Privacy-Preserving Collaborative Machine Learning

Dual-Cloud Multi-Secret Sharing Architecture for Privacy Preserving Persistent Computation

LEGO: A Hybrid Toolkit for Efficient 2Pc-Based Privacy-Preserving Machine Learning

Privacy-Preserving Distributed Machine Learning Made Faster.

Customizing Trusted AI Accelerators for Efficient Privacy-Preserving Machine Learning

When approximate design for fast homomorphic computation provides differential privacy guarantees

A Multi-Layer Parallel Hardware Architecture for Homomorphic Computation in Machine Learning

SecureML: A System for Scalable Privacy-Preserving Machine Learning

A Survey of Deep Learning Architectures for Privacy-Preserving Machine Learning With Fully Homomorphic Encryption

Privacy Preservation for Machine Learning Training and Classification Based on Homomorphic Encryption Schemes

A Pervasive, Efficient and Private Future: Realizing Privacy-Preserving Machine Learning Through Hybrid Homomorphic Encryption