10), and the noise can be treated as a perturbation. We study optical neural networks where all layers except the last are operated in the limit that each neuron can be activated by just a single photon, and as a result the noise on neuron activations is no longer merely perturbative. We show that by using a physics-based probabilistic model of the neuron activations in training, it is possible to perform accurate machine-learning inference in spite of the extremely high shot noise (SNR ~ 1). We experimentally demonstrated MNIST handwritten-digit classification with a test accuracy of 98% using an optical neural network with a hidden layer operating in the single-photon regime; the optical energy used to perform the classification corresponds to just 0.038 photons per multiply-accumulate (MAC) operation. Our physics-aware stochastic training approach might also prove useful with non-optical ultra-low-power hardware."> 10), and the noise can be treated as a perturbation. We study optical neural networks where all layers except the last are operated in the limit that each neuron can be activated by just a single photon, and as a result the noise on neuron activations is no longer merely perturbative. We show that by using a physics-based probabilistic model of the neuron activations in training, it is possible to perform accurate machine-learning inference in spite of the extremely high shot noise (SNR ~ 1). We experimentally demonstrated MNIST handwritten-digit classification with a test accuracy of 98% using an optical neural network with a hidden layer operating in the single-photon regime; the optical energy used to perform the classification corresponds to just 0.038 photons per multiply-accumulate (MAC) operation. Our physics-aware stochastic training approach might also prove useful with non-optical ultra-low-power hardware.">
lynx   »   [go: up one dir, main page]
IDEAS home Printed from https://ideas.repec.org/a/nat/natcom/v16y2025i1d10.1038_s41467-024-55220-y.html
   My bibliography  Save this article

Quantum-limited stochastic optical neural networks operating at a few quanta per activation

Author

Listed:
  • Shi-Yuan Ma

    (School of Applied and Engineering Physics, Cornell University)

  • Tianyu Wang

    (School of Applied and Engineering Physics, Cornell University)

  • Jérémie Laydevant

    (School of Applied and Engineering Physics, Cornell University
    USRA Research Institute for Advanced Computer Science)

  • Logan G. Wright

    (School of Applied and Engineering Physics, Cornell University
    NTT Research, Inc.
    Yale University)

  • Peter L. McMahon

    (School of Applied and Engineering Physics, Cornell University
    Cornell University)

Abstract
Energy efficiency in computation is ultimately limited by noise, with quantum limits setting the fundamental noise floor. Analog physical neural networks hold promise for improved energy efficiency compared to digital electronic neural networks. However, they are typically operated in a relatively high-power regime so that the signal-to-noise ratio (SNR) is large (>10), and the noise can be treated as a perturbation. We study optical neural networks where all layers except the last are operated in the limit that each neuron can be activated by just a single photon, and as a result the noise on neuron activations is no longer merely perturbative. We show that by using a physics-based probabilistic model of the neuron activations in training, it is possible to perform accurate machine-learning inference in spite of the extremely high shot noise (SNR ~ 1). We experimentally demonstrated MNIST handwritten-digit classification with a test accuracy of 98% using an optical neural network with a hidden layer operating in the single-photon regime; the optical energy used to perform the classification corresponds to just 0.038 photons per multiply-accumulate (MAC) operation. Our physics-aware stochastic training approach might also prove useful with non-optical ultra-low-power hardware.

Suggested Citation

  • Shi-Yuan Ma & Tianyu Wang & Jérémie Laydevant & Logan G. Wright & Peter L. McMahon, 2025. "Quantum-limited stochastic optical neural networks operating at a few quanta per activation," Nature Communications, Nature, vol. 16(1), pages 1-12, December.
    • Handle: RePEc:nat:natcom:v:16:y:2025:i:1:d:10.1038_s41467-024-55220-y
    DOI: 10.1038/s41467-024-55220-y
    as

    Download full text from publisher

    File URL: https://www.nature.com/articles/s41467-024-55220-y
    File Function: Abstract
    Download Restriction: no
    File URL: https://libkey.io/10.1038/s41467-024-55220-y?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    References listed on IDEAS

    as
    1. Vinay Joshi & Manuel Le Gallo & Simon Haefeli & Irem Boybat & S. R. Nandakumar & Christophe Piveteau & Martino Dazzi & Bipin Rajendran & Abu Sebastian & Evangelos Eleftheriou, 2020. "Accurate deep neural network inference using computational phase-change memory," Nature Communications, Nature, vol. 11(1), pages 1-13, December.
    2. M. Prezioso & F. Merrikh-Bayat & B. D. Hoskins & G. C. Adam & K. K. Likharev & D. B. Strukov, 2015. "Training and operation of an integrated neuromorphic network based on metal-oxide memristors," Nature, Nature, vol. 521(7550), pages 61-64, May.
    3. Tianyu Wang & Shi-Yuan Ma & Logan G. Wright & Tatsuhiro Onodera & Brian C. Richard & Peter L. McMahon, 2022. "An optical neural network using less than 1 photon per multiplication," Nature Communications, Nature, vol. 13(1), pages 1-8, December.
    4. Jacob Torrejon & Mathieu Riou & Flavio Abreu Araujo & Sumito Tsunegi & Guru Khalsa & Damien Querlioz & Paolo Bortolotti & Vincent Cros & Kay Yakushiji & Akio Fukushima & Hitoshi Kubota & Shinji Yuasa , 2017. "Neuromorphic computing with nanoscale spintronic oscillators," Nature, Nature, vol. 547(7664), pages 428-431, July.
    5. Wim Bogaerts & Daniel Pérez & José Capmany & David A. B. Miller & Joyce Poon & Dirk Englund & Francesco Morichetti & Andrea Melloni, 2020. "Programmable photonic circuits," Nature, Nature, vol. 586(7828), pages 207-216, October.
    6. Tao Chen & Jeroen van Gelder & Bram van de Ven & Sergey V. Amitonov & Bram de Wilde & Hans-Christian Ruiz Euler & Hajo Broersma & Peter A. Bobbert & Floris A. Zwanenburg & Wilfred G. van der Wiel, 2020. "Classification with a disordered dopant-atom network in silicon," Nature, Nature, vol. 577(7790), pages 341-345, January.
    7. J. Feldmann & N. Youngblood & M. Karpov & H. Gehring & X. Li & M. Stappers & M. Gallo & X. Fu & A. Lukashchuk & A. S. Raja & J. Liu & C. D. Wright & A. Sebastian & T. J. Kippenberg & W. H. P. Pernice , 2021. "Parallel convolutional processing using an integrated photonic tensor core," Nature, Nature, vol. 589(7840), pages 52-58, January.
    8. Logan G. Wright & Tatsuhiro Onodera & Martin M. Stein & Tianyu Wang & Darren T. Schachter & Zoey Hu & Peter L. McMahon, 2022. "Deep physical neural networks trained with backpropagation," Nature, Nature, vol. 601(7894), pages 549-555, January.
    9. J. Feldmann & N. Youngblood & M. Karpov & H. Gehring & X. Li & M. Stappers & M. Gallo & X. Fu & A. Lukashchuk & A. S. Raja & J. Liu & C. D. Wright & A. Sebastian & T. J. Kippenberg & W. H. P. Pernice , 2021. "Publisher Correction: Parallel convolutional processing using an integrated photonic tensor core," Nature, Nature, vol. 591(7849), pages 13-13, March.
    10. Xingyuan Xu & Mengxi Tan & Bill Corcoran & Jiayang Wu & Andreas Boes & Thach G. Nguyen & Sai T. Chu & Brent E. Little & Damien G. Hicks & Roberto Morandotti & Arnan Mitchell & David J. Moss, 2021. "11 TOPS photonic convolutional accelerator for optical neural networks," Nature, Nature, vol. 589(7840), pages 44-51, January.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Mitsumasa Nakajima & Katsuma Inoue & Kenji Tanaka & Yasuo Kuniyoshi & Toshikazu Hashimoto & Kohei Nakajima, 2022. "Physical deep learning with biologically inspired training method: gradient-free approach for physical hardware," Nature Communications, Nature, vol. 13(1), pages 1-12, December.
    2. Ziyu Zhan & Hao Wang & Qiang Liu & Xing Fu, 2024. "Photonic diffractive generators through sampling noises from scattering media," Nature Communications, Nature, vol. 15(1), pages 1-9, December.
    3. Alexander Song & Sai Nikhilesh Murty Kottapalli & Rahul Goyal & Bernhard Schölkopf & Peer Fischer, 2024. "Low-power scalable multilayer optoelectronic neural networks enabled with incoherent light," Nature Communications, Nature, vol. 15(1), pages 1-12, December.
    4. Han Zhao & Bingzhao Li & Huan Li & Mo Li, 2022. "Enabling scalable optical computing in synthetic frequency dimension using integrated cavity acousto-optics," Nature Communications, Nature, vol. 13(1), pages 1-7, December.
    5. Yunping Bai & Yifu Xu & Shifan Chen & Xiaotian Zhu & Shuai Wang & Sirui Huang & Yuhang Song & Yixuan Zheng & Zhihui Liu & Sim Tan & Roberto Morandotti & Sai T. Chu & Brent E. Little & David J. Moss & , 2025. "TOPS-speed complex-valued convolutional accelerator for feature extraction and inference," Nature Communications, Nature, vol. 16(1), pages 1-13, December.
    6. Wen Zhou & Bowei Dong & Nikolaos Farmakidis & Xuan Li & Nathan Youngblood & Kairan Huang & Yuhan He & C. David Wright & Wolfram H. P. Pernice & Harish Bhaskaran, 2023. "In-memory photonic dot-product engine with electrically programmable weight banks," Nature Communications, Nature, vol. 14(1), pages 1-10, December.
    7. Wenting Wang & Ping-Keng Lu & Abhinav Kumar Vinod & Deniz Turan & James F. McMillan & Hao Liu & Mingbin Yu & Dim-Lee Kwong & Mona Jarrahi & Chee Wei Wong, 2022. "Coherent terahertz radiation with 2.8-octave tunability through chip-scale photomixed microresonator optical parametric oscillation," Nature Communications, Nature, vol. 13(1), pages 1-9, December.
    8. H. H. Zhu & J. Zou & H. Zhang & Y. Z. Shi & S. B. Luo & N. Wang & H. Cai & L. X. Wan & B. Wang & X. D. Jiang & J. Thompson & X. S. Luo & X. H. Zhou & L. M. Xiao & W. Huang & L. Patrick & M. Gu & L. C., 2022. "Space-efficient optical computing with an integrated chip diffractive neural network," Nature Communications, Nature, vol. 13(1), pages 1-9, December.
    9. Dongliang Wang & Yikun Nie & Gaolei Hu & Hon Ki Tsang & Chaoran Huang, 2024. "Ultrafast silicon photonic reservoir computing engine delivering over 200 TOPS," Nature Communications, Nature, vol. 15(1), pages 1-11, December.
    10. Bowen Bai & Qipeng Yang & Haowen Shu & Lin Chang & Fenghe Yang & Bitao Shen & Zihan Tao & Jing Wang & Shaofu Xu & Weiqiang Xie & Weiwen Zou & Weiwei Hu & John E. Bowers & Xingjun Wang, 2023. "Microcomb-based integrated photonic processing unit," Nature Communications, Nature, vol. 14(1), pages 1-10, December.
    11. Chenduan Chen & Zhan Yang & Tao Wang & Yalun Wang & Kai Gao & Jiajia Wu & Jun Wang & Jianrong Qiu & Dezhi Tan, 2024. "Ultra-broadband all-optical nonlinear activation function enabled by MoTe2/optical waveguide integrated devices," Nature Communications, Nature, vol. 15(1), pages 1-12, December.
    12. Xiangyan Meng & Guojie Zhang & Nuannuan Shi & Guangyi Li & José Azaña & José Capmany & Jianping Yao & Yichen Shen & Wei Li & Ninghua Zhu & Ming Li, 2023. "Compact optical convolution processing unit based on multimode interference," Nature Communications, Nature, vol. 14(1), pages 1-9, December.
    13. Xuguang Zhang & Zixuan Zhou & Yijun Guo & Minxue Zhuang & Warren Jin & Bitao Shen & Yujun Chen & Jiahui Huang & Zihan Tao & Ming Jin & Ruixuan Chen & Zhangfeng Ge & Zhou Fang & Ning Zhang & Yadong Liu, 2024. "High-coherence parallelization in integrated photonics," Nature Communications, Nature, vol. 15(1), pages 1-9, December.
    14. Miltiadis Moralis-Pegios & George Giamougiannis & Apostolos Tsakyridis & David Lazovsky & Nikos Pleros, 2024. "Perfect linear optics using silicon photonics," Nature Communications, Nature, vol. 15(1), pages 1-8, December.
    15. Guangwei Cong & Noritsugu Yamamoto & Takashi Inoue & Yuriko Maegami & Morifumi Ohno & Shota Kita & Shu Namiki & Koji Yamada, 2022. "On-chip bacterial foraging training in silicon photonic circuits for projection-enabled nonlinear classification," Nature Communications, Nature, vol. 13(1), pages 1-12, December.
    16. Steven Becker & Dirk Englund & Birgit Stiller, 2024. "An optoacoustic field-programmable perceptron for recurrent neural networks," Nature Communications, Nature, vol. 15(1), pages 1-8, December.
    17. Cansu Demirkiran & Lakshmi Nair & Darius Bunandar & Ajay Joshi, 2024. "A blueprint for precise and fault-tolerant analog neural networks," Nature Communications, Nature, vol. 15(1), pages 1-14, December.
    18. Maoliang Wei & Kai Xu & Bo Tang & Junying Li & Yiting Yun & Peng Zhang & Yingchun Wu & Kangjian Bao & Kunhao Lei & Zequn Chen & Hui Ma & Chunlei Sun & Ruonan Liu & Ming Li & Lan Li & Hongtao Lin, 2024. "Monolithic back-end-of-line integration of phase change materials into foundry-manufactured silicon photonics," Nature Communications, Nature, vol. 15(1), pages 1-9, December.
    19. Liuting Shan & Qizhen Chen & Rengjian Yu & Changsong Gao & Lujian Liu & Tailiang Guo & Huipeng Chen, 2023. "A sensory memory processing system with multi-wavelength synaptic-polychromatic light emission for multi-modal information recognition," Nature Communications, Nature, vol. 14(1), pages 1-11, December.
    20. G. Mourgias-Alexandris & M. Moralis-Pegios & A. Tsakyridis & S. Simos & G. Dabos & A. Totovic & N. Passalis & M. Kirtas & T. Rutirawut & F. Y. Gardes & A. Tefas & N. Pleros, 2022. "Noise-resilient and high-speed deep learning with coherent silicon photonics," Nature Communications, Nature, vol. 13(1), pages 1-7, December.

    More about this item

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:nat:natcom:v:16:y:2025:i:1:d:10.1038_s41467-024-55220-y. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Sonal Shukla or Springer Nature Abstracting and Indexing (email available below). General contact details of provider: http://www.nature.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.
    Лучший частный хостинг