Photonic Differential Privacy with Direct Feedback Alignment
This work provides a private-by-design training solution for deep learning using photonic chips, addressing privacy concerns in machine learning applications.
The authors tackled the problem of training deep neural networks with privacy guarantees by leveraging the intrinsic noise of optical random projections to create a differentially private Direct Feedback Alignment mechanism, achieving solid end-task performance in experiments.
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.