Scaling up for end-to-end on-chip photonic neural network inference
Open Access
- 17 September 2025
- journal article
- research article
- Published by Springer Nature in Light: Science & Applications
- Vol. 14 (1) , 1-10
Abstract
Optical neural networks are emerging as a competitive alternative to their electronic counterparts, offering distinct advantages in bandwidth and energy efficiency. Despite these benefits, scaling up on-chip optical neural networks for end-to-end inference is facing significant challenges. First, network depth is constrained by the weak cascadability of optical nonlinear activation functions. Second, the input size is constrained by the scale of the optical matrix. Herein, we propose a scaling up strategy called partially coherent deep optical neural networks (PDONNs). By leveraging an on-chip nonlinear activation function based on opto-electro-opto conversion, PDONN enables network depth expansion with positive net gain. Additionally, convolutional layers achieve rapid dimensionality reduction, thereby allowing for an increase in the accommodated input size. The use of a partially coherent optical source significantly reduces reliance on narrow-linewidth laser diodes and coherent detection. Owing to their broader spectral characteristics and simpler implementation, such sources are more accessible and compatible with scalable integration. Benefiting from these innovations, we designed and fabricated a monolithically integrated optical neural network with the largest input size and the deepest network depth, comprising an input layer with a size of 64, two convolutional layers, and two fully connected layers. We successfully demonstrate end-to-end two-class classification of fashion images and four-class classification of handwritten digits with accuracies of 96% and 94%, respectively, using an in-situ training method. Notably, performance is well maintained with partially coherent illumination. This proposed architecture represents a critical step toward realizing energy-efficient, scalable, and widely accessible optical computing.Keywords
This publication has 41 references indexed in Scilit:
- Experimentally realized in situ backpropagation for deep learning in photonic neural networksScience, 2023
- Photonic machine learning with on-chip diffractive opticsNature Communications, 2023
- Chip-Based High-Dimensional Optical Neural NetworkNano-Micro Letters, 2022
- Nonlinear germanium-silicon photodiode for activation and monitoring in photonic neuromorphic networksNature Communications, 2022
- An on-chip photonic deep neural network for image classificationNature, 2022
- Photonic matrix multiplication lights up photonic accelerator and beyondLight: Science & Applications, 2022
- An optical neural chip for implementing complex-valued neural networkNature Communications, 2021
- Parallel convolutional processing using an integrated photonic tensor coreNature, 2021
- All-optical spiking neurosynaptic networks with self-learning capabilitiesNature, 2019
- All-optical machine learning using diffractive deep neural networksScience, 2018