Optical coherent dot-product chip for sophisticated deep learning regression

Shaofu Xu; Jing Wang; Haowen Shu; Zhike Zhang; Sicheng Yi; Bowen Bai; Xingjun Wang; Jianguo Liu; Weiwen Zou

doi:10.1038/s41377-021-00666-8

Go to Nature Website

Your Location：

Home >

Browse articles >

Optical coherent dot-product chip for sophisticated deep learning regression

Articles

- Optical coherent dot-product chip for sophisticated deep learning regression
- Light: Science & Applications Vol. 10, Issue 12, Pages: 2330-2341(2021)
- Affiliations：
  
  1.State Key Laboratory of Advanced Optical Communication Systems and Networks, Intelligent Microwave Lightwave Integration Innovation Center (imLic), Department of Electronic Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China
  2.State Key Laboratory of Advanced Optical Communications System and Networks, Department of Electronics, School of Electronics Engineering and Computer Science, Peking University, Beijing 100871, China
  3.Institution of Semiconductors, Chinese Academy of Sciences, Beijing 100083, China
- Author bio：
  
  Weiwen Zou (wzou@sjtu.edu.cn)
- Funds：
- DOI：10.1038/s41377-021-00666-8
  CLC：
- Published：31 December 2021，
  
  Published Online：01 November 2021，
  
  Received：28 May 2021，
  
  Revised：29 September 2021，
  
  Accepted：18 October 2021
- Accepted：
Scan QR Code
Xu, S. F. et al. Optical coherent dot-product chip for sophisticated deep learning regression. Light: Science & Applications, 10, 2330-2341 (2021).
DOI：

Xu, S. F. et al. Optical coherent dot-product chip for sophisticated deep learning regression. Light: Science & Applications, 10, 2330-2341 (2021). DOI： 10.1038/s41377-021-00666-8.

摘要

Abstract

Optical implementations of neural networks (ONNs) herald the next-generation high-speed and energy-efficient deep learning computing by harnessing the technical advantages of large bandwidth and high parallelism of optics. However

due to the problems of the incomplete numerical domain

limited hardware scale

or inadequate numerical accuracy

the majority of existing ONNs were studied for basic classification tasks. Given that regression is a fundamental form of deep learning and accounts for a large part of current artificial intelligence applications

it is necessary to master deep learning regression for further development and deployment of ONNs. Here

we demonstrate a silicon-based optical coherent dot-product chip (OCDC) capable of completing deep learning regression tasks. The OCDC adopts optical fields to carry out operations in the complete real-value domain instead of in only the positive domain. Via reusing

a single chip conducts matrix multiplications and convolutions in neural networks of any complexity. Also

hardware deviations are compensated via in-situ backpropagation control provided the simplicity of chip architecture. Therefore

the OCDC meets the requirements for sophisticated regression tasks and we successfully demonstrate a representative neural network

the AUTOMAP (a cutting-edge neural network model for image reconstruction). The quality of reconstructed images by the OCDC and a 32-bit digital computer is comparable. To the best of our knowledge

there is no precedent of performing such state-of-the-art regression tasks on ONN chips. It is anticipated that the OCDC can promote the novel accomplishment of ONNs in modern AI applications including autonomous driving

natural language processing

and scientific study.

关键词

Keywords

references

LeCun, Y., Bengio, Y.&Hinton, G. Deep learning.Nature521, 436–444 (2015)..

Grigorescu, S. et al. A survey of deep learning techniques for autonomous driving.J. Field Robot.37, 362–386 (2020)..

Chen, H. et al. Low-dose CT with a residual encoder-decoder convolutional neural network.IEEE Trans. Med. Imaging36, 2524–2535 (2017)..

Segler, M. H. S., Preuss, M.&Waller, M. P. Planning chemical syntheses with deep neural networks and symbolic AI.Nature555, 604–610 (2018)..

Deringer, V. L. et al. Origins of structural and electronic transitions in disordered silicon.Nature589, 59–64 (2021)..

Xu, S. F. et al. Deep-learning-powered photonic analog-to-digital conversion.Light8, 66 (2019)..

Senior, A. W. et al. Improved protein structure prediction using potentials from deep learning.Nature577, 706–710 (2020)..

Waldrop, M. M. The chips are down for Moore's law.Nature530, 144–147 (2016)..

Amodei, D.&Hernandez, D.AI and Computehttps://openai.com/blog/ai-and-compute/#addendumhttps://openai.com/blog/ai-and-compute/#addendum(2018)..

Shastri, B. J. et al. Photonics for artificial intelligence and neuromorphic computing.Nat. Photonics15, 102–114(2021)..

Nahmias, M. A. et al. Photonic multiply-accumulate operations for neural networks.IEEE J. Sel. Top. Quantum Electron.26, 7701518 (2020)..

Xu, S. F., Wang, J.&Zou, W. W. Optical convolutional neural network with WDM-based optical patching and microring weighting banks.IEEE Photonics Technol. Lett.33, 89–92 (2021)..

Totović, A. R. et al. Femtojoule per MAC neuromorphic photonics: an energy and technology roadmap.IEEE J. Sel. Top. Quantum Electron.26, 8800115 (2020)..

Chen, Z. G.&Segev, M. Highlighting photonics: looking into the next decade.eLight1, 2 (2021)..

Shen, Y. C. et al. Deep learning with coherent nanophotonic circuits.Nat. Photonics11, 441–446 (2017)..

Tait, A. N. et al. Neuromorphic photonic networks using silicon photonic weight banks.Sci. Rep.7, 7430 (2017)..

Zhou, T. K. et al. Large-scale neuromorphic optoelectronic computing with a reconfigurable diffractive processing unit.Nat. Photonics15, 367–373 (2021)..

Xu, S. F. et al. High-accuracy optical convolution unit architecture for convolutional neural networks by cascaded acousto-optical modulator arrays.Opt. Express27, 19778–19787 (2019)..

Hamerly, R. et al. Large-scale optical neural networks based on photoelectric multiplication.Phys. Rev. X9, 021032 (2019)..

Huang, Y. Y. et al. Programmable matrix operation with reconfigurable time-wavelength plane manipulation and dispersed time delay.Opt. Express27, 20456–20467 (2019)..

Xu, X. Y. et al. 11 TOPS photonic convolutional accelerator for optical neural networks.Nature589, 44–51 (2021)..

Feldmann, J. et al. Parallel convolutional processing using an integrated photonic tensor core.Nature589, 52–58 (2021)..

LeCun, Y., Bottou, L., Bengio, Y.&Haffner, P. Gradient-based learning applied to document recognition.Proc. IEEE86, 2278–2324 (1998)..

Rivenson, Y. et al. Phase recovery and holographic image reconstruction using deep learning in neural networks.Light7, 17141 (2018)..

Jaderberg, M. et al. Human-level performance in 3D multiplayer games with population-based reinforcement learning.Science364, 859–865 (2019)..

Liu, D. J. et al. Training deep neural networks for the inverse design of nanophotonic structures.ACS Photonics5, 1365–1369 (2018)..

Peurifoy, J. et al. Nanophotonic particle simulation and inverse design using artificial neural networks.Sci. Adv.4, eaar4206 (2018)..

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

Mourgias-Alexandris, G. et al. Neuromorphic photonics with coherent linear neurons using dual-IQ modulation cells.J. Lightwave Technol.38, 811–819 (2020)..

Zhu, B. et al. Image reconstruction by domain-transform manifold learning.Nature555, 487–492 (2018)..

Liu, P. F., Qiu, X. P.&Huang, X. J. Recurrent neural network for text classification with multi-task learning.Proc. 25th International Joint Conference on Artificial Intelligence. (ACM, New York, 2016) 2873–2879.

Ronneberger, O., Fischer, P.&Brox, T. U-Net: convolutional networks for biomedical image segmentation.Proc. 18th International Conference on Medical Image Computing and Computer-Assisted Intervention. (Springer, Munich, 2015) 234–241.

Chetlur, S. et al. cuDNN: efficient primitives for deep learning. Preprint athttps://arxiv.org/abs/1410.0759https://arxiv.org/abs/1410.0759(2014).

Xu, S. F., Wang, J.&Zou, W. W. Optical patching scheme for optical convolutional neural networks based on wavelength-division multiplexing and optical delay lines.Opt. Lett.45, 3689–3692 (2020)..

Brown, R. W. et al.Magnetic Resonance Imaging: Physical Principles and Sequence Design. 2nd edn. (John Wiley&Sons, Inc., Hoboken, 2014).

Lustig, M., Donoho, D.&Pauly, J. M. Sparse MRI: the application of compressed sensing for rapid MR imaging.Magn. Reson. Med.58, 1182–1195 (2007)..

Lindgren, A. G.&Rattey, P. A. The inverse discrete Radon transform with applications to tomographic imaging using projection data.Adv. Electron. Electron Phys.56, 359–410 (1981)..

Wang, C. et al. Integrated lithium niobate electro-optic modulators operating at CMOS-compatible voltages.Nature562, 101–104 (2018)..

Weigel, P. O. et al. Bonded thin film lithium niobate modulator on a silicon photonics platform exceeding 100 GHz 3-dB electrical modulation bandwidth.Opt. Express26, 23728–23739 (2018)..

He, M. B. et al. High-performance hybrid silicon and lithium niobate Mach–Zehnder modulators for 100 Gbit s-1 and beyond.Nat. Photonics13, 359–364 (2019)..

Poot, M.&Tang, H. X. Broadband nanoelectromechanical phase shifting of light on a chip.Appl. Phys. Lett.104, 061101 (2014)..

Sun, C. et al. Single-chip microprocessor that communicates directly using light.Nature528, 534–538 (2015)..

Atabaki, A. H. et al. Integratingphotonics with silicon nanoelectronics for the next generation of systems on a chip.Nature556, 349–354 (2018)..

Fan, Q. Y. et al. MGH–USC Human Connectome Project datasets with ultra-high b-value diffusion MRI.Neuroimage124, 1108–1114 (2016)..

Kingma, D. P.&Ba, J.Adam: A Method for Stochastic Optimization. Preprint athttps://arxiv.org/abs/1412.6980https://arxiv.org/abs/1412.6980(2014)..

Hughes, T. W. et al. Training of photonic neural networks through in situ backpropagation and gradient measurement.Optica5, 864–871 (2018)..

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