Photonic Spiking Neural Networks and Graphene-on-Silicon Spiking Neurons

Spiking neural networks are known to be superior over artificial neural networks for their computational power efficiency and noise robustness. The benefits of spiking coupled with the high-bandwidth and low-latency of photonics can enable highly-efficient, noise-robust, high-speed neural processors. The landscape of photonic spiking neurons consists of an overwhelming majority of excitable lasers and a few demonstrations on nonlinear optical cavities. The silicon platform is best poised to host a scalable photonic technology given its CMOS-compatibility and low optical loss. Here, we present a survey of existing photonic spiking neurons, and propose a novel spiking neuron based on a hybrid graphene-on-silicon microring cavity. A comparison among a representative sample of photonic spiking devices is also presented. Finally, we discuss methods employed in training spiking neural networks, their challenges as well as the application domain that can be enabled by photonic spiking neural hardware.

[1]  L. Pesquera,et al.  Nonlinear dynamics induced by parallel and orthogonal optical injection in 1550 nm Vertical-Cavity Surface-Emitting Lasers (VCSELs). , 2010, Optics express.

[2]  Bernabé Linares-Barranco,et al.  Mapping from Frame-Driven to Frame-Free Event-Driven Vision Systems by Low-Rate Rate Coding and Coincidence Processing--Application to Feedforward ConvNets , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  C. Wright,et al.  Photonics for artificial intelligence and neuromorphic computing , 2020, ArXiv.

[4]  Pierre Tirilly,et al.  Unsupervised Visual Feature Learning with Spike-timing-dependent Plasticity: How Far are we from Traditional Feature Learning Approaches? , 2019, Pattern Recognit..

[5]  Kaushik Roy,et al.  Going Deeper in Spiking Neural Networks: VGG and Residual Architectures , 2018, Front. Neurosci..

[6]  E. Kriezis,et al.  Rigorous calculation of nonlinear parameters in graphene-comprising waveguides , 2015 .

[7]  Darpan T. Sanghavi,et al.  BindsNET: A Machine Learning-Oriented Spiking Neural Networks Library in Python , 2018, Front. Neuroinform..

[8]  R Kuszelewicz,et al.  Relative refractory period in an excitable semiconductor laser. , 2014, Physical review letters.

[9]  Eugene M. Izhikevich,et al.  Resonate-and-fire neurons , 2001, Neural Networks.

[10]  Frank C. Hoppensteadt,et al.  Bursts as a unit of neural information: selective communication via resonance , 2003, Trends in Neurosciences.

[11]  Shaowu Chen,et al.  Bistability and self-pulsation phenomena in silicon microring resonators based on nonlinear optical effects. , 2012, Optics express.

[12]  Sylvain Barbay,et al.  Excitability in a semiconductor laser with saturable absorber. , 2011, Optics letters.

[13]  Rodrigo Alvarez-Icaza,et al.  Neurogrid: A Mixed-Analog-Digital Multichip System for Large-Scale Neural Simulations , 2014, Proceedings of the IEEE.

[14]  Hong Wang,et al.  Loihi: A Neuromorphic Manycore Processor with On-Chip Learning , 2018, IEEE Micro.

[15]  Matěj Hejda,et al.  Neuromorphic encoding of image pixel data into rate-coded optical spike trains with a photonic VCSEL-neuron , 2021 .

[16]  Patrick D. Wolf,et al.  Evaluation of spike-detection algorithms fora brain-machine interface application , 2004, IEEE Transactions on Biomedical Engineering.

[17]  Xiaohua Li,et al.  Temporal Pulses Driven Spiking Neural Network for Fast Object Recognition in Autonomous Driving , 2020, ArXiv.

[18]  Ting Wang,et al.  Enhanced optical Kerr nonlinearity of graphene/Si hybrid waveguide , 2018, 2018 Asia Communications and Photonics Conference (ACP).

[19]  James G. Mitchell,et al.  Flip-chip integrated silicon photonic bridge chips for sub-picojoule per bit optical links , 2010, 2010 Proceedings 60th Electronic Components and Technology Conference (ECTC).

[20]  Kaushik Roy,et al.  Towards spike-based machine intelligence with neuromorphic computing , 2019, Nature.

[21]  Mable P. Fok,et al.  Photonic implementation of a neuronal algorithm applicable towards angle of arrival detection and localization. , 2015, Optics express.

[22]  Pieter G. Kik,et al.  Erbium-Doped Optical-Waveguide Amplifiers on Silicon , 1998 .

[23]  Wolfgang Maass,et al.  Lower Bounds for the Computational Power of Networks of Spiking Neurons , 1996, Neural Computation.

[24]  Daan Lenstra,et al.  Multipulse excitability in a semiconductor laser with optical injection. , 2002, Physical review letters.

[25]  Paul R. Prucnal,et al.  Simulations of a graphene excitable laser for spike processing , 2014 .

[26]  Paul R. Prucnal,et al.  Spike processing with a graphene excitable laser , 2016, Scientific Reports.

[27]  M Radziunas,et al.  Excitability of a semiconductor laser by a two-mode homoclinic bifurcation. , 2001, Physical review letters.

[28]  Bipin Rajendran,et al.  NormAD - Normalized Approximate Descent based supervised learning rule for spiking neurons , 2015, 2015 International Joint Conference on Neural Networks (IJCNN).

[29]  Yue Hao,et al.  Computing Primitive of Fully VCSEL-Based All-Optical Spiking Neural Network for Supervised Learning and Pattern Classification , 2020, IEEE Transactions on Neural Networks and Learning Systems.

[30]  Salvador Balle,et al.  Experimental evidence of van der Pol-Fitzhugh-Nagumo dynamics in semiconductor optical amplifiers. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[31]  Joni Dambre,et al.  Excitability in optically injected microdisk lasers with phase controlled excitatory and inhibitory response. , 2013, Optics express.

[32]  M. Mohrle,et al.  Gigahertz self-pulsation in 1.5 mu m wavelength multisection DFB lasers , 1992, IEEE Photonics Technology Letters.

[33]  Miguel A. Larotonda,et al.  Experimental investigation on excitability in a laser with a saturable absorber , 2002 .

[34]  N. Matsuda,et al.  Optical nonlinearity enhancement with graphene-decorated silicon waveguides , 2017, Scientific Reports.

[35]  B. Schrauwen,et al.  Cascadable excitability in microrings. , 2012, Optics express.

[36]  Jianye Zhao,et al.  Optical spike-timing-dependent plasticity with weight-dependent learning window and reward modulation. , 2015, Optics express.

[37]  I. Sagnes,et al.  Excitability and self-pulsing in a photonic crystal nanocavity , 2012 .

[38]  Lei Deng,et al.  Spatio-Temporal Backpropagation for Training High-Performance Spiking Neural Networks , 2017, Front. Neurosci..

[39]  Paul R. Prucnal,et al.  A Laser Spiking Neuron in a Photonic Integrated Circuit. , 2020, 2012.08516.

[40]  Rüdiger Paschotta,et al.  Experimentally confirmed design guidelines for passively Q-switched microchip lasers using semiconductor saturable absorbers , 2001 .

[41]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[42]  Susumu Noda,et al.  Investigation of optical nonlinearities in an ultra-high-Q Si nanocavity in a two-dimensional photonic crystal slab. , 2006, Optics express.

[43]  Salvador Balle,et al.  Excitability and optical pulse generation in semiconductor lasers driven by resonant tunneling diode photo-detectors. , 2013, Optics express.

[44]  S. Massar,et al.  Measuring the nonlinear refractive index of graphene using the optical Kerr effect method. , 2016, Optics letters.

[45]  P. Prucnal,et al.  High-speed all-optical thresholding via carrier lifetime tunability. , 2020, Optics letters.

[46]  Shuiying Xiang,et al.  STDP-Based Unsupervised Spike Pattern Learning in a Photonic Spiking Neural Network With VCSELs and VCSOAs , 2019, IEEE Journal of Selected Topics in Quantum Electronics.

[47]  Henry Markram,et al.  On the computational power of circuits of spiking neurons , 2004, J. Comput. Syst. Sci..

[48]  Mario Dagenais,et al.  Optical injection induced polarization bistability in vertical‐cavity surface‐emitting lasers , 1993 .

[49]  Andrzej J. Kasinski,et al.  Supervised Learning in Spiking Neural Networks with ReSuMe: Sequence Learning, Classification, and Spike Shifting , 2010, Neural Computation.

[50]  A. Marini,et al.  Theory of graphene saturable absorption , 2016, 1605.06499.

[51]  B Kelleher,et al.  Excitability in optically injected semiconductor lasers: contrasting quantum-well- and quantum-dot-based devices. , 2011, Physical review. E, Statistical, nonlinear, and soft matter physics.

[52]  Tobi Delbrück,et al.  Training Deep Spiking Neural Networks Using Backpropagation , 2016, Front. Neurosci..

[53]  Yue Hao,et al.  Numerical Implementation of Wavelength-Dependent Photonic Spike Timing Dependent Plasticity Based on VCSOA , 2018, IEEE Journal of Quantum Electronics.

[54]  A. M. van der Zande,et al.  Regenerative oscillation and four-wave mixing in graphene optoelectronics , 2012, Conference on Lasers and Electro-Optics.

[55]  D Goulding,et al.  Excitability in a quantum dot semiconductor laser with optical injection. , 2007, Physical review letters.

[56]  B Krauskopf,et al.  Excitability and coherence resonance in lasers with saturable absorber. , 1999, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[57]  Paul R. Prucnal,et al.  Recent progress in semiconductor excitable lasers for photonic spike processing , 2016 .

[58]  J. Feldmann,et al.  All-optical spiking neurosynaptic networks with self-learning capabilities , 2019, Nature.

[59]  Sebastian Wieczorek,et al.  Excitability and self-pulsations near homoclinic bifurcations in semiconductor laser systems , 2003 .

[60]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[61]  G. Agrawal,et al.  Nonlinear optical phenomena in silicon waveguides: modeling and applications. , 2007, Optics express.

[62]  Michael Pfeiffer,et al.  Deep Learning With Spiking Neurons: Opportunities and Challenges , 2018, Front. Neurosci..

[63]  Vasileios G. Ataloglou,et al.  Nonlinear coupled-mode-theory framework for graphene-induced saturable absorption in nanophotonic resonant structures , 2018, Physical Review A.

[64]  P. R. Prucnal,et al.  A Leaky Integrate-and-Fire Laser Neuron for Ultrafast Cognitive Computing , 2013, IEEE Journal of Selected Topics in Quantum Electronics.

[65]  Yue Tian,et al.  Pulse lead/lag timing detection for adaptive feedback and control based on optical spike-timing-dependent plasticity. , 2013, Optics letters.

[66]  Tobi Delbrück,et al.  A Low Power, Fully Event-Based Gesture Recognition System , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[67]  Adonis Bogris,et al.  Micro-Ring-Resonator Based Passive Photonic Spike-Time-Dependent-Plasticity Scheme for Unsupervised Learning in Optical Neural Networks , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[68]  J. Danckaert,et al.  Excitability in semiconductor microring lasers: Experimental and theoretical pulse characterization , 2010, 1108.3704.

[69]  Bhavin J. Shastri,et al.  Neuromorphic Photonic Integrated Circuits , 2018, IEEE Journal of Selected Topics in Quantum Electronics.

[70]  Indranil Chakraborty,et al.  Toward Fast Neural Computing using All-Photonic Phase Change Spiking Neurons , 2018, Scientific Reports.

[71]  Indranil Chakraborty,et al.  Photonic In-Memory Computing Primitive for Spiking Neural Networks Using Phase-Change Materials , 2019, Physical Review Applied.

[72]  Genquan Han,et al.  Spike Sequence Learning in a Photonic Spiking Neural Network Consisting of VCSELs-SA With Supervised Training , 2020, IEEE Journal of Selected Topics in Quantum Electronics.

[73]  Paul R. Prucnal,et al.  Temporal Information Processing With an Integrated Laser Neuron , 2020, IEEE Journal of Selected Topics in Quantum Electronics.

[74]  Wolfgang Maass,et al.  Networks of Spiking Neurons: The Third Generation of Neural Network Models , 1996, Electron. Colloquium Comput. Complex..

[75]  Bernard Brezzo,et al.  TrueNorth: Design and Tool Flow of a 65 mW 1 Million Neuron Programmable Neurosynaptic Chip , 2015, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[76]  J. Danckaert,et al.  Optical injection in semiconductor ring lasers , 2009 .

[77]  B. Krauskopf,et al.  Self-pulsations of lasers with saturable absorber: dynamics and bifurcations , 1999 .

[78]  H. Sompolinsky,et al.  The tempotron: a neuron that learns spike timing–based decisions , 2006, Nature Neuroscience.

[79]  Ricardo Lent.  Resource Selection in Cognitive Networks With Spiking Neural Networks , 2018, IEEE Transactions on Cognitive Communications and Networking.

[80]  H. Thienpont,et al.  Negative Kerr nonlinearity of graphene as seen via chirped-pulse-pumped self-phase modulation , 2016, 1611.07750.

[81]  R. Soref,et al.  Electrooptical effects in silicon , 1987 .

[82]  P. Prucnal,et al.  NEUROMORPHIC PHOTONICS , 2017 .

[83]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[84]  Antonio Hurtado,et al.  Investigation of vertical cavity surface emitting laser dynamics for neuromorphic photonic systems , 2012 .

[85]  Matthew Cook,et al.  Fast-classifying, high-accuracy spiking deep networks through weight and threshold balancing , 2015, 2015 International Joint Conference on Neural Networks (IJCNN).

[86]  P. Monnier,et al.  Fast thermo-optical excitability in a two-dimensional photonic crystal. , 2006, Physical review letters.

[87]  Sander M. Bohte,et al.  SpikeProp: backpropagation for networks of spiking neurons , 2000, ESANN.