Scaling up for end-to-end on-chip photonic neural network inference

Bo Wu; Chaoran Huang; Jialong Zhang; Hailong Zhou; Yilun Wang; Jianji Dong; Xinliang Zhang

doi:10.1038/s41377-025-02029-z

Your Location：

Home >

Browse articles >

Scaling up for end-to-end on-chip photonic neural network inference

Original Articles

- Scaling up for end-to-end on-chip photonic neural network inference
- Light: Science & Applications Vol. 14, Issue 11, Pages: 3526-3535(2025)
- Affiliations：
  
  1.Wuhan National Laboratory for Optoelectronics, School of Optical and Electronic Information, Huazhong University of Science and Technology, Wuhan 430074, China
  2.Department of Electronic Engineering, The Chinese University of Hong Kong, Shatin, Hong Kong SAR, China
- Author bio：
  
  Hailong Zhou (hailongzhou@hust.edu.cn)
  Jianji Dong (jjdong@hust.edu.cn)
- Funds：
- DOI：10.1038/s41377-025-02029-z
  CLC：
- Received：13 April 2025，
  
  Revised：2025-07-24，
  
  Accepted：31 August 2025，
  
  Published Online：17 September 2025，
  
  Published：30 November 2025
- Accepted：
Scan QR Code
Wu, B. et al. Scaling up for end-to-end on-chip photonic neural network inference. Light: Science & Applications, 14, 3526-3535 (2025).
DOI：

Wu, B. et al. Scaling up for end-to-end on-chip photonic neural network inference. Light: Science & Applications, 14, 3526-3535 (2025). DOI： 10.1038/s41377-025-02029-z.

摘要

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.

关键词

Keywords

references

Zhang, W. Q. et al. Neuro-inspired computing chips. Nat. Electron. 3 , 371–382 (2020)..

Fu, T. Z. et al. Optical neural networks: progress and challenges. Light Sci. Appl. 13 , 263 (2024)..

Shekhar, S. et al. Roadmapping the next generation of silicon photonics. Nat. Commun. 15 , 751 (2024)..

Zhang, W. K. et al. Large-scale optical programmable logic array for two-dimensional cellular automaton. Adv. Photonics 6 , 056007 (2024)..

Ashtiani, F., Geers, A. J. & Aflatouni, F. An on-chip photonic deep neural network for image classification. Nature 606 , 501–506 (2022)..

Bandyopadhyay, S. et al. Single-chip photonic deep neural network with forward-only training. Nat. Photonics 18 , 1335–1343 (2024)..

Zhou, H. L. et al. Photonic matrix multiplication lights up photonic accelerator and beyond. Light Sci. Appl. 11 , 30 (2022)..

Zhou, T. K. et al. Large-scale neuromorphic optoelectronic computing with a reconfigurable diffractive processing unit. Nat. Photonics 15 , 367–373 (2021)..

Lin, X. et al. All-optical machine learning using diffractive deep neural networks. Science 361 , 1004–1008 (2018)..

Chen, Y. T. et al. All-analog photoelectronic chip for high-speed vision tasks. Nature 623 , 48–57 (2023)..

Wu, B. et al. Chip-to-chip optical multimode communication with universal mode processors. PhotoniX 4 , 37 (2023)..

Xu, Z. H. et al. Large-scale photonic chiplet Taichi empowers 160-TOPS/W artificial general intelligence. Science 384 , 202–209 (2024)..

Du, Z. C. et al. Ultracompact and multifunctional integrated photonic platform. Sci. Adv. 10 , eadm7569 (2024)..

Pai, S. et al. Experimentally realized in situ backpropagation for deep learning in photonic neural networks. Science 380 , 398–404 (2023)..

Wu, B. et al. Monolithically integrated asynchronous optical recurrent accelerator. eLight 5 , 7 (2025)..

Zhang, W. P. et al. A system-on-chip microwave photonic processor solves dynamic RF interference in real time with picosecond latency. Light Sci. Appl. 13 , 14 (2024)..

Huang, C. R. et al. A silicon photonic-electronic neural network for fibre nonlinearity compensation. Nat. Electron. 4 , 837–844 (2021)..

Feldmann, J. et al. Parallel convolutional processing using an integrated photonic tensor core. Nature 589 , 52–58 (2021)..

Dong, B. W. et al. Higher-dimensional processing using a photonic tensor core with continuous-time data. Nat. Photonics 17 , 1080–1088 (2023)..

Cheng, J. W. et al. Multimodal deep learning using on-chip diffractive optics with in situ training capability. Nat. Commun. 15 , 6189 (2024)..

Fu, T. Z. et al. P hotonic machine learning with on-chip diffractive optics. Nat. Commun. 14 , 70 (2023)..

Zhang, J. L. et al. Highly integrated all-optical nonlinear deep neural network for multi-thread processing. Adv. Photonics 7 , 046003(2025)..

Rahim, A. et al. Taking silicon photonics modulators to a higher performance level: state-of-the-art and a review of new technologies. Adv. Photonics 3 , 024003 (2021)..

Shen, Y. C. et al. Deep learning with coherent nanophotonic circuits. Nat. Photonics 11 , 441–446 (2017)..

Wu, B. et al. Programmable integrated photonic coherent matrix: principle, configuring, and applications. Appl. Phys. Rev. 11 , 011309 (2024)..

Wu, B. et al. Low-threshold all-optical nonlinear activation function based on a Ge/Si hybrid structure in a microring resonator. Opt. Mater. Express 12 , 970–980 (2022)..

Shi, Y. et al. Nonlinear germanium-silicon photodiode for activation and monitoring in photonic neuromorphic networks. Nat. Commun. 13 , 6048 (2022)..

Wang, T. Y. et al. Image sensing with multilayer nonlinear optical neural networks. Nat. Photonics 17 , 408–415 (2023)..

Huang, Z. et al. Pre-sensor computing with compact multilayer optical neural network. Sci. Adv. 10 , eado8516 (2024)..

Zhao, B. H. et al. Cascadable optical nonlinear activation function based on Ge–Si. Opt. Lett. 49 , 6149–6152 (2024)..

Zhang, H. et al. An optical neural chip for implementing complex-valued neural network. Nat. Commun. 12 , 457 (2021)..

Dong, B. W. et al. Partial coherence enhances parallelized photonic computing. Nature 632 , 55–62 (2024)..

Tait, A. N. et al. Silicon photonic modulator neuron. Phys. Rev. Appl. 11 , 064043 (2019)..

Tang, R. et al. Waveguide-multiplexed photonic matrix-vector multiplication processor using multiport photodetectors. Optica 12 , 812–820 (2025)..

Wu, B. et al. Real-valued optical matrix computing with simplified MZI mesh. Intell. Comput. 2 , 0047 (2023)..

Wang, X. Y. et al. Chip-based high-dimensional optical neural network. Nano Micro Lett. 14 , 221 (2022)..

Rizzo, A. et al. Massively scalable Kerr comb-driven silicon photonic link. Nat. Photonics 17 , 781–790 (2023)..

Zhang, J. H. et al. Compact, efficient, and scalable nanobeam core for photonic matrix-vector multiplication. Optica 11 , 190–196 (2024)..

Nozaki, K. et al. Femtofarad optoelectronic integration demonstrating energy-saving signal conversion and nonlinear functions. Nat. Photonics 13 , 454–459 (2019)..

McMahon, P. L. The physics of optical computing. Nat. Rev. Phys. 5 , 717–734 (2023)..

Kingma, D. P. & Ba, J. Adam: a method for stochastic optimization. In Proc. 3rd International Conference on Learning Representations (San Diego, 2015).

Feldmann, J. et al. All-optical spiking neurosynaptic networks with self-learning capabilities. Nature 569 , 208–214 (2019)..

Views

Downloads

CSCD

Alert me when the article has been cited

Submit

Tools

Publicity Resources

No data

Related Author

No data

Related Institution

No data

⁰