Programmable phase-change metasurfaces on waveguides for multimode photonic convolutional neural network

Abstract

Neuromorphic photonics has recently emerged as a promising hardware accelerator, with significant potential speed and energy advantages over digital electronics for machine learning algorithms, such as neural networks of various types. Integrated photonic networks are particularly powerful in performing analog computing of matrix-vector multiplication (MVM) as they afford unparalleled speed and bandwidth density for data transmission. Incorporating nonvolatile phase-change materials in integrated photonic devices enables indispensable programming and in-memory computing capabilities for on-chip optical computing. Here, we demonstrate a multimode photonic computing core consisting of an array of programable mode converters based on on-waveguide metasurfaces made of phase-change materials. The programmable converters utilize the refractive index change of the phase-change material Ge 2 Sb 2 Te 5 during phase transition to control the waveguide spatial modes with a very high precision of up to 64 levels in modal contrast. This contrast is used to represent the matrix elements, with 6-bit resolution and both positive and negative values, to perform MVM computation in neural network algorithms. We demonstrate a prototypical optical convolutional neural network that can perform image processing and recognition tasks with high accuracy. With a broad operation bandwidth and a compact device footprint, the demonstrated multimode photonic core is promising toward large-scale photonic neural networks with ultrahigh computation throughputs.

Full text

Its licence does not allow us to show the full text. Read it at PubMed Central:

References 65

1. 1.

   Marr, B., Degnan, B., Hasler, P. & Anderson, D. Scaling Energy Per Operation via an Asynchronous Pipeline. IEEE Transactions on Very Large Scale. Integration (VLSI) Systems. 21, 147–151 (IEEE, 2013).

   Not in our library

2. 2.

   Jones N. How to stop data centres from gobbling up the world’s electricity. Nature. 2018;561:163-166.

   Not in our library10.1038/d41586-018-06610-y (opens in a new tab)

3. 3.

   Athale R, Psaltis D. Optical computing: past and future. Opt. Photon. News. 2016;27:32-39.

   Not in our library10.1364/opn.27.6.000032 (opens in a new tab)

4. 4.

   Solli DR, Jalali B. Analog optical computing. Nat. Photon. 2015;9:704-706.

   Not in our library10.1038/nphoton.2015.208 (opens in a new tab)

5. 5.

   Prucnal, P. R. & Shastri, B. J. Neuromorphic Photonics (CRC Press, 2017).

   Not in our library

6. 6.

   Caulfield HJ, Dolev S. Why future supercomputing requires optics. Nat. Photon. 2010;4:261-263.

   Not in our library10.1038/nphoton.2010.94 (opens in a new tab)

7. 7.

   Zhang C, Zhang S, Peters JD, Bowers JE. 8 × 8 × 40 Gbps fully integrated silicon photonic network on chip. Optica. 2016;3:785-786.

   Not in our library10.1364/optica.3.000785 (opens in a new tab)

8. 8.

   Shen YW, et al. Silicon photonics for extreme scale systems. J. Lightwave Technol. 2019;37:245-259.

   Not in our library10.1109/jlt.2019.2897365 (opens in a new tab)

9. 9.

   Wade, M. et al. In 2018 European Conference on Optical Communication (ECOC). 1–3 (IEEE, 2018).

   Not in our library

10. 10.

    Wuttig M, Bhaskaran H, Taubner T. Phase-change materials for non-volatile photonic applications. Nat. Photon. 2017;11:465-476.

    Not in our library10.1038/nphoton.2017.126 (opens in a new tab)

11. 11.

    Yang Z, Ramanathan S. Breakthroughs in photonics 2014: phase change materials for photonics. IEEE Photon. J. 2015;7:1-5.

    Not in our library

12. 12.

    Zhang W, Mazzarello R, Wuttig M, Ma E. Designing crystallization in phase-change materials for universal memory and neuro-inspired computing. Nat. Rev. Mater. 2019;4:150-168.

    Not in our library10.1038/s41578-018-0076-x (opens in a new tab)

13. 13.

    Briggs RM, Pryce IM, Atwater HA. Compact silicon photonic waveguide modulator based on the vanadium dioxide metal-insulator phase transition. Opt. Express. 2010;18:11192-11201.

    Not in our library10.1364/oe.18.011192 (opens in a new tab)

14. 14.

    Wang Q, et al. Optically reconfigurable metasurfaces and photonic devices based on phase change materials. Nat. Photon. 2016;10:60-U75.

    Not in our library10.1038/nphoton.2015.247 (opens in a new tab)

15. 15.

    Chu CH, et al. Active dielectric metasurface based on phase‐change medium. Laser Photon. Rev. 2016;10:986-994.

    Not in our library10.1002/lpor.201600106 (opens in a new tab)

16. 16.

    Yin X, et al. Beam switching and bifocal zoom lensing using active plasmonic metasurfaces. Light. Sci. Appl. 2017;6:e17016.

    Not in our library10.1038/lsa.2017.16 (opens in a new tab)

17. 17.

    Wu C, et al. Low-loss integrated photonic switch using subwavelength patterned phase change material. ACS Photon. 2018;6:87-92.

    Not in our library10.1021/acsphotonics.8b01516 (opens in a new tab)

18. 18.

    Cheng Z, et al. Device-level photonic memories and logic applications using phase-change materials. Adv. Mater. 2018;30:e1802435.

    Not in our library10.1002/adma.201802435 (opens in a new tab)

19. 19.

    Xu P, Zheng J, Doylend JK, Majumdar A. Low-loss and broadband nonvolatile phase-change directional coupler switches. ACS Photon. 2019;6:553-557.

    Not in our library10.1021/acsphotonics.8b01628 (opens in a new tab)

20. 20.
    Broadband transparent optical phase change materials for high-performance nonvolatile photonics · 2019

    No notice
21. 21.

    de Galarreta CR, et al. Nonvolatile reconfigurable phase-change metadevices for beam steering in the near infrared. Adv. Funct. Mater. 2018;28:1704993.

    Not in our library10.1002/adfm.201704993 (opens in a new tab)

22. 22.

    Stegmaier M, Ríos C, Bhaskaran H, Wright CD, Pernice WHP. Nonvolatile all-optical 1 × 2 switch for chipscale photonic networks. Adv. Opt. Mater. 2017;5:1600346.

    Not in our library10.1002/adom.201600346 (opens in a new tab)

23. 23.

    Zhang Q, et al. Broadband nonvolatile photonic switching based on optical phase change materials: beyond the classical figure-of-merit. Opt. Lett. 2018;43:94.

    Not in our library10.1364/ol.43.000094 (opens in a new tab)

24. 24.

    Li X, et al. Fast and reliable storage using a 5-bit, non-volatile photonic memory cell. Optica. 2019;6:1-6.

    Not in our library10.1364/optica.6.000001 (opens in a new tab)

25. 25.

    Martins T, et al. Fiber-integrated phase-change reconfigurable optical attenuator. Appl. Photon. 2019;4:111301.

    Not in our library10.1063/1.5116000 (opens in a new tab)

26. 26.

    Zheng J, et al. Nonvolatile electrically reconfigurable integrated photonic switch enabled by a silicon PIN diode heater. Adv. Mater. 2020;32:e2001218.

    Not in our library10.1002/adma.202001218 (opens in a new tab)

27. 27.

    George JK, et al. Neuromorphic photonics with electro-absorption modulators. Opt. Express. 2019;27:5181-5191.

    Not in our library10.1364/oe.27.005181 (opens in a new tab)

28. 28.

    Tait AN, et al. Silicon photonic modulator neuron. Phys. Rev. Appl. 2019;11:064043.

    Not in our library10.1103/physrevapplied.11.064043 (opens in a new tab)

29. 29.

    Hamerly R, Bernstein L, Sludds A, Soljacic M, Englund D. Large-scale optical neural networks based on photoelectric multiplication. Phys. Rev. X. 2019;9:021032.

    Not in our library

30. 30.

    Shen Y, et al. Deep learning with coherent nanophotonic circuits. Nat. Photon. 2017;11:441-446.

    Not in our library10.1038/nphoton.2017.93 (opens in a new tab)

31. 31.

    Sun J, Timurdogan E, Yaacobi A, Hosseini ES, Watts MR. Large-scale nanophotonic phased array. Nature. 2013;493:195-199.

    Not in our library10.1038/nature11727 (opens in a new tab)

32. 32.

    Ribeiro A, Ruocco A, Vanacker L, Bogaerts W. Demonstration of a 4x4-port universal linear circuit. Optica. 2016;3:1348-1357.

    Not in our library10.1364/optica.3.001348 (opens in a new tab)

33. 33.

    Bocker RP. Matrix multiplication using incoherent optical techniques. Appl. Opt. 1974;13:1670-1676.

    Not in our library10.1364/ao.13.001670 (opens in a new tab)

34. 34.

    Ríos C, et al. In-memory computing on a photonic platform. Sci. Adv. 2019;5:eaau5759.

    Not in our library10.1126/sciadv.aau5759 (opens in a new tab)

35. 35.

    Chakraborty I, Saha G, Roy K. Photonic in-memory computing primitive for spiking neural networks using phase-change materials. Phys. Rev. Appl. 2019;11:014063.

    Not in our library10.1103/physrevapplied.11.014063 (opens in a new tab)

36. 36.

    Caulfield HJ, Kinser J, Rogers SK. Optical neural networks. Proc. IEEE. 1989;77:1573-1583.

    Not in our library10.1109/5.40669 (opens in a new tab)

37. 37.

    Feldmann, J. et al. Parallel convolution processing using an integrated photonic tensor core. arXiv preprint arXiv:2002.00281 (2020).

    Not in our library

38. 38.

    Feldmann J, Youngblood N, Wright CD, Bhaskaran H, Pernice WHP. All-optical spiking neurosynaptic networks with self-learning capabilities. Nature. 2019;569:208-214.

    Not in our library10.1038/s41586-019-1157-8 (opens in a new tab)

39. 39.

    Li Z, et al. Controlling propagation and coupling of waveguide modes using phase-gradient metasurfaces. Nat. Nanotechnol. 2017;12:675-683.

    Not in our library10.1038/nnano.2017.50 (opens in a new tab)

40. 40.

    Park J-W, et al. Optical properties of pseudobinary GeTe, Ge2Sb2Te5, GeSb2Te4, GeSb4Te7, and Sb2Te3 from ellipsometry and density functional theory. Phys. Rev. B. 2009;80:115209.

    Not in our library10.1103/physrevb.80.115209 (opens in a new tab)

41. 41.

    Liu, Y., Aziz, M. M., Shalini, A., Wright, C. D. & Hicken, R. J. Crystallization of Ge2Sb2Te5 films by amplified femtosecond optical pulses. J. Appl. Phys.112, 123526 (2012).

    Not in our library

42. 42.

    Farmakidis N, et al. Plasmonic nanogap enhanced phase-change devices with dual electrical-optical functionality. Sci. Adv. 2019;5:eaaw2687.

    Not in our library10.1126/sciadv.aaw2687 (opens in a new tab)

43. 43.

    Zhang H, et al. Miniature multilevel optical memristive switch using phase change material. ACS Photon. 2019;6:2205-2212.

    Not in our library10.1021/acsphotonics.9b00819 (opens in a new tab)

44. 44.

    Rodriguez-Hernandez G, Hosseini P, Ríos C, Wright CD, Bhaskaran H. Mixed-mode electro-optical operation of Ge2Sb2Te5 nanoscale crossbar devices. Adv. Electron. Mater. 2017;3:1700079.

    Not in our library10.1002/aelm.201700079 (opens in a new tab)

45. 45.

    Ríos C, et al. Integrated all-photonic non-volatile multi-level memory. Nat. Photon. 2015;9:725-732.

    Not in our library10.1038/nphoton.2015.182 (opens in a new tab)

46. 46.

    Giannopoulos, I. et al. In 2018 IEEE International Electron Devices Meeting (IEDM). 27.27.21–27.27.24 (IEEE, 2018).

    Not in our library

47. 47.

    Le Gallo M, et al. Mixed-precision in-memory computing. Nat. Electron. 2018;1:246-253.

    Not in our library10.1038/s41928-018-0054-8 (opens in a new tab)

48. 48.

    Nahmias MA, et al. Photonic multiply-accumulate operations for neural networks. IEEE J. Sel. Top. Quant. 2020;26:1-18.

    Not in our library10.1109/jstqe.2019.2941485 (opens in a new tab)

49. 49.

    Convolution Neural Network - simple code - simple to use (MATLAB Central File Exchange, 2020).

    Not in our library

50. 50.

    Xiong C, et al. Monolithic 56 Gb/s silicon photonic pulse-amplitude modulation transmitter. Optica. 2016;3:1060-1065.

    Not in our library10.1364/optica.3.001060 (opens in a new tab)

51. 51.

    Moazeni S, et al. A 40-Gb/s PAM-4 Transmitter Based on a Ring-Resonator Optical DAC in 45-nm SOI CMOS. IEEE J. Solid-State Circuits. 2017;52:3503-3516.

    Not in our library10.1109/jssc.2017.2748620 (opens in a new tab)

52. 52.

    Sawchuk, A. A. & Jenkins, B. K. In Optical Computing. 143–153 (International Society for Optics and Photonics 1986).

    Not in our library

53. 53.

    Joshi, A. et al. In 2009 3rd ACM/IEEE International Symposium on Networks-on-Chip. 124-133 (IEEE, 2009).

    Not in our library

54. 54.

    Khope ASP, et al. Multi-wavelength selective crossbar switch. Opt. Express. 2019;27:5203-5216.

    Not in our library10.1364/oe.27.005203 (opens in a new tab)

55. 55.

    Ohno S, Toprasertpong K, Takagi S, Takenaka M. Si microring resonator crossbar arrays for deep learning accelerator. Jpn J. Appl. Phys. 2020;59:SGGE04.

    Not in our library10.35848/1347-4065/ab6d82 (opens in a new tab)

56. 56.

    Nahmias MA, et al. Photonic multiply-accumulate operations for neural networks. IEEE J. Sel. Top. Quant. Electron. 2019;26:1-18.

    Not in our library10.1109/jstqe.2019.2941485 (opens in a new tab)

57. 57.

    Han S, Seok TJ, Quack N, Yoo B-W, Wu MC. Large-scale silicon photonic switches with movable directional couplers. Optica. 2015;2:370-375.

    Not in our library10.1364/optica.2.000370 (opens in a new tab)

58. 58.

    Tait AN, Chang J, Shastri BJ, Nahmias MA, Prucnal PR. Demonstration of WDM weighted addition for principal component analysis. Opt. Express. 2015;23:12758.

    Not in our library10.1364/oe.23.012758 (opens in a new tab)

59. 59.

    Tait AN, Nahmias MA, Shastri BJ, Prucnal PR. Broadcast and weight: an integrated network for scalable photonic spike processing. J. Lightwave Technol. 2014;32:4029-4041.

    Not in our library10.1109/jlt.2014.2345652 (opens in a new tab)

60. 60.

    Silver D, et al. A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play. Science. 2018;362:1140-114.

    Not in our library10.1126/science.aar6404 (opens in a new tab)

61. 61.

    Jouppi, N. P. et al. In Proceedings of the 44th Annual International Symposium on Computer Architecture - ISCA ‘17 1-12 (ACM Press, New York, New York, USA, 2017).

    Not in our library

62. 62.

    Li X, et al. Experimental investigation of silicon and silicon nitride platforms for phase-change photonic in-memory computing. Optica. 2020;7:218-225.

    Not in our library10.1364/optica.379228 (opens in a new tab)

63. 63.

    Gayen DK, Chattopadhyay T, Pal RK, Roy JN. All-optical Multiplication with the help of Semiconductor Optical Amplifier—assisted Sagnac Switch. J. Comput. Electron. 2010;9:57-67.

    Not in our library10.1007/s10825-010-0305-z (opens in a new tab)

64. 64.

    Atabaki AH, et al. Integrating photonics with silicon nanoelectronics for the next generation of systems on a chip. Nature. 2018;556:349-354.

    Not in our library10.1038/s41586-018-0028-z (opens in a new tab)

65. 65.

    Bangari V, et al. Digital electronics and analog photonics for convolutional neural networks (DEAP-CNNs). IEEE J. Sel. Top. Quant. 2020;26:1-13.

    Not in our library10.1109/jstqe.2019.2945540 (opens in a new tab)

Cited by papers we hold 3

• On-chip bacterial foraging training in silicon photonic circuits for projection-enabled nonlinear classification · 2022
  Not checked
• The challenges of modern computing and new opportunities for optics · 2021
  Not checked
• Optical meta-waveguides for integrated photonics and beyond · 2021
  No notice

Questions about this paper

Reads

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.