IDEAS home Printed from https://ideas.repec.org/a/nat/nature/v601y2022i7894d10.1038_s41586-021-04223-6.html
   My bibliography  Save this article

Deep physical neural networks trained with backpropagation

Author

Listed:
  • Logan G. Wright

    (Cornell University
    NTT Research, Inc.)

  • Tatsuhiro Onodera

    (Cornell University
    NTT Research, Inc.)

  • Martin M. Stein

    (Cornell University)

  • Tianyu Wang

    (Cornell University)

  • Darren T. Schachter

    (Cornell University)

  • Zoey Hu

    (Cornell University)

  • Peter L. McMahon

    (Cornell University)

Abstract

Deep-learning models have become pervasive tools in science and engineering. However, their energy requirements now increasingly limit their scalability1. Deep-learning accelerators2–9 aim to perform deep learning energy-efficiently, usually targeting the inference phase and often by exploiting physical substrates beyond conventional electronics. Approaches so far10–22 have been unable to apply the backpropagation algorithm to train unconventional novel hardware in situ. The advantages of backpropagation have made it the de facto training method for large-scale neural networks, so this deficiency constitutes a major impediment. Here we introduce a hybrid in situ–in silico algorithm, called physics-aware training, that applies backpropagation to train controllable physical systems. Just as deep learning realizes computations with deep neural networks made from layers of mathematical functions, our approach allows us to train deep physical neural networks made from layers of controllable physical systems, even when the physical layers lack any mathematical isomorphism to conventional artificial neural network layers. To demonstrate the universality of our approach, we train diverse physical neural networks based on optics, mechanics and electronics to experimentally perform audio and image classification tasks. Physics-aware training combines the scalability of backpropagation with the automatic mitigation of imperfections and noise achievable with in situ algorithms. Physical neural networks have the potential to perform machine learning faster and more energy-efficiently than conventional electronic processors and, more broadly, can endow physical systems with automatically designed physical functionalities, for example, for robotics23–26, materials27–29 and smart sensors30–32.

Suggested Citation

  • 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.
  • Handle: RePEc:nat:nature:v:601:y:2022:i:7894:d:10.1038_s41586-021-04223-6
    DOI: 10.1038/s41586-021-04223-6
    as

    Download full text from publisher

    File URL: https://www.nature.com/articles/s41586-021-04223-6
    File Function: Abstract
    Download Restriction: Access to the full text of the articles in this series is restricted.

    File URL: https://libkey.io/10.1038/s41586-021-04223-6?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
    ---><---

    As the access to this document is restricted, you may want to search for a different version of it.

    Citations

    Citations are extracted by the CitEc Project, subscribe to its RSS feed for this item.
    as


    Cited by:

    1. Kilian D. Stenning & Jack C. Gartside & Luca Manneschi & Christopher T. S. Cheung & Tony Chen & Alex Vanstone & Jake Love & Holly Holder & Francesco Caravelli & Hidekazu Kurebayashi & Karin Everschor-, 2024. "Neuromorphic overparameterisation and few-shot learning in multilayer physical neural networks," Nature Communications, Nature, vol. 15(1), pages 1-15, December.
    2. Hao He & Maofeng Cao & Yun Gao & Peng Zheng & Sen Yan & Jin-Hui Zhong & Lei Wang & Dayong Jin & Bin Ren, 2024. "Noise learning of instruments for high-contrast, high-resolution and fast hyperspectral microscopy and nanoscopy," Nature Communications, Nature, vol. 15(1), pages 1-12, December.
    3. Chengkuan Gao & Prabhav Gaur & Dhaifallah Almutairi & Shimon Rubin & Yeshaiahu Fainman, 2023. "Optofluidic memory and self-induced nonlinear optical phase change for reservoir computing in silicon photonics," Nature Communications, Nature, vol. 14(1), pages 1-9, December.
    4. Abbas, Khizar & Han, Mengyao & Xu, Deyi & Butt, Khalid Manzoor & Baz, Khan & Cheng, Jinhua & Zhu, Yongguang & Hussain, Sanwal, 2024. "Exploring synergistic and individual causal effects of rare earth elements and renewable energy on multidimensional economic complexity for sustainable economic development," Applied Energy, Elsevier, vol. 364(C).
    5. Seou Choi & Yannick Salamin & Charles Roques-Carmes & Rumen Dangovski & Di Luo & Zhuo Chen & Michael Horodynski & Jamison Sloan & Shiekh Zia Uddin & Marin Soljačić, 2024. "Photonic probabilistic machine learning using quantum vacuum noise," Nature Communications, Nature, vol. 15(1), pages 1-8, December.
    6. Jamshaid Ul Rahman & Sana Danish & Dianchen Lu, 2023. "Deep Neural Network-Based Simulation of Sel’kov Model in Glycolysis: A Comprehensive Analysis," Mathematics, MDPI, vol. 11(14), pages 1-9, July.
    7. Fan Cai & Yuesong Jiang & Wanqing Song & Kai-Hung Lu & Tongbo Zhu, 2024. "Short-Term Wind Turbine Blade Icing Wind Power Prediction Based on PCA-fLsm," Energies, MDPI, vol. 17(6), pages 1-15, March.
    8. Fangjun Hu & Saeed A. Khan & Nicholas T. Bronn & Gerasimos Angelatos & Graham E. Rowlands & Guilhem J. Ribeill & Hakan E. Türeci, 2024. "Overcoming the coherence time barrier in quantum machine learning on temporal data," Nature Communications, Nature, vol. 15(1), pages 1-12, December.
    9. Jérémie Laydevant & Danijela Marković & Julie Grollier, 2024. "Training an Ising machine with equilibrium propagation," Nature Communications, Nature, vol. 15(1), pages 1-14, December.
    10. Ruomin Zhu & Sam Lilak & Alon Loeffler & Joseph Lizier & Adam Stieg & James Gimzewski & Zdenka Kuncic, 2023. "Online dynamical learning and sequence memory with neuromorphic nanowire networks," Nature Communications, Nature, vol. 14(1), pages 1-12, December.
    11. Federico Ricci & Massimiliano Avana & Francesco Mariani, 2024. "Enhancing Lambda Measurement in Hydrogen-Fueled SI Engines through Virtual Sensor Implementation," Energies, MDPI, vol. 17(16), pages 1-17, August.
    12. Gao Wang & Giulia Marcucci & Benjamin Peters & Maria Chiara Braidotti & Lars Muckli & Daniele Faccio, 2024. "Human-centred physical neuromorphics with visual brain-computer interfaces," Nature Communications, Nature, vol. 15(1), pages 1-8, December.
    13. 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.
    14. Sunkyu Yu & Namkyoo Park, 2023. "Heavy tails and pruning in programmable photonic circuits for universal unitaries," Nature Communications, Nature, vol. 14(1), pages 1-10, December.
    15. Tianyu Wang & Jialin Meng & Xufeng Zhou & Yue Liu & Zhenyu He & Qi Han & Qingxuan Li & Jiajie Yu & Zhenhai Li & Yongkai Liu & Hao Zhu & Qingqing Sun & David Wei Zhang & Peining Chen & Huisheng Peng & , 2022. "Reconfigurable neuromorphic memristor network for ultralow-power smart textile electronics," Nature Communications, Nature, vol. 13(1), pages 1-8, December.
    16. Md Tauhidul Islam & Zixia Zhou & Hongyi Ren & Masoud Badiei Khuzani & Daniel Kapp & James Zou & Lu Tian & Joseph C. Liao & Lei Xing, 2023. "Revealing hidden patterns in deep neural network feature space continuum via manifold learning," Nature Communications, Nature, vol. 14(1), pages 1-20, December.
    17. Malte J. Rasch & Charles Mackin & Manuel Gallo & An Chen & Andrea Fasoli & Frédéric Odermatt & Ning Li & S. R. Nandakumar & Pritish Narayanan & Hsinyu Tsai & Geoffrey W. Burr & Abu Sebastian & Vijay N, 2023. "Hardware-aware training for large-scale and diverse deep learning inference workloads using in-memory computing-based accelerators," Nature Communications, Nature, vol. 14(1), pages 1-18, December.
    18. Hongmei Cui & Zhongyang Li & Bingchuan Sun & Teng Fan & Yonghao Li & Lida Luo & Yong Zhang & Jian Wang, 2022. "A New Ice Quality Prediction Method of Wind Turbine Impeller Based on the Deep Neural Network," Energies, MDPI, vol. 15(22), pages 1-18, November.

    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:nature:v:601:y:2022:i:7894:d:10.1038_s41586-021-04223-6. 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.

    We have no bibliographic references for this item. You can help adding them by using 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.