Photonic Differential Privacy with Direct Feedback Alignment

TL;DR

This paper introduces a novel differentially private training method for deep neural networks using optical processing units' intrinsic noise, enabling privacy-preserving learning with theoretical guarantees and solid empirical results.

Contribution

It leverages optical noise in OPUs to create a privacy mechanism for DFA, combining theoretical analysis with practical experiments.

Findings

Achieves differential privacy through optical noise in OPUs.

Provides theoretical bounds on privacy guarantees.

Demonstrates competitive performance on neural network training.

Abstract

Optical Processing Units (OPUs) -- low-power photonic chips dedicated to large scale random projections -- have been used in previous work to train deep neural networks using Direct Feedback Alignment (DFA), an effective alternative to backpropagation. Here, we demonstrate how to leverage the intrinsic noise of optical random projections to build a differentially private DFA mechanism, making OPUs a solution of choice to provide a private-by-design training. We provide a theoretical analysis of our adaptive privacy mechanism, carefully measuring how the noise of optical random projections propagates in the process and gives rise to provable Differential Privacy. Finally, we conduct experiments demonstrating the ability of our learning procedure to achieve solid end-task performance.