Fully Homomorphically Encrypted Deep Learning as a Service

TL;DR

This paper explores the application of Fully Homomorphic Encryption in deep learning, demonstrating its potential for privacy-preserving, scalable, and reasonably efficient predictions, exemplified by milk yield forecasting.

Contribution

It derives and proves methods for implementing FHE with deep learning at scale, addressing associated challenges and proposing mitigations for practical deployment.

Findings

FHE enables private deep learning predictions with acceptable time complexity.

Spatial complexity of FHE is high, but manageable for practical applications.

FHE can facilitate secure data sharing in supply chains, like in agriculture.

Abstract

Fully Homomorphic Encryption (FHE) is a relatively recent advancement in the field of privacy-preserving technologies. FHE allows for the arbitrary depth computation of both addition and multiplication, and thus the application of abelian/polynomial equations, like those found in deep learning algorithms. This project investigates, derives, and proves how FHE with deep learning can be used at scale, with relatively low time complexity, the problems that such a system incurs, and mitigations/solutions for such problems. In addition, we discuss how this could have an impact on the future of data privacy and how it can enable data sharing across various actors in the agri-food supply chain, hence allowing the development of machine learning-based systems. Finally, we find that although FHE incurs a high spatial complexity cost, the time complexity is within expected reasonable bounds, while allowing for absolutely private predictions to be made, in our case for milk yield prediction.