Asymmetrical estimator for training encapsulated deep photonic neural networks

TL;DR

This paper introduces Asymmetrical Training (AsyT), a novel method for efficiently training encapsulated deep photonic neural networks (DPNNs) that maintains analogue signal propagation, reducing computational overhead and system complexity.

Contribution

The paper presents AsyT, a lightweight, error-tolerant training approach specifically designed for encapsulated DPNNs, enabling faster, energy-efficient operation with minimal readouts despite fabrication variations.

Findings

AsyT improves training efficiency over traditional backpropagation methods.

The method demonstrates consistent performance enhancement across different network structures.

AsyT reduces system footprint and energy consumption in photonic neural network training.

Abstract

Photonic neural networks (PNNs) are fast in-propagation and high bandwidth paradigms that aim to popularize reproducible NN acceleration with higher efficiency and lower cost. However, the training of PNN is known to be challenging, where the device-to-device and system-to-system variations create imperfect knowledge of the PNN. Despite backpropagation (BP)-based training algorithms being the industry standard for their robustness, generality, and fast gradient convergence for digital training, existing PNN-BP methods rely heavily on accurate intermediate state extraction or extensive computational resources for deep PNNs (DPNNs). The truncated photonic signal propagation and the computation overhead bottleneck DPNN's operation efficiency and increase system construction cost. Here, we introduce the asymmetrical training (AsyT) method, tailored for encapsulated DPNNs, where the signal is preserved in the analogue photonic domain for the entire structure. AsyT offers a lightweight solution for DPNNs with minimum readouts, fast and energy-efficient operation, and minimum system footprint. AsyT's ease of operation, error tolerance, and generality aim to promote PNN acceleration in a widened operational scenario despite the fabrication variations and imperfect controls. We demonstrated AsyT for encapsulated DPNN with integrated photonic chips, repeatably enhancing the performance from in-silico BP for different network structures and datasets.