Training on the optical system: local search method

Optical neural networks hold tremendous potential for energy efficiency and low latency. Despite this potential, the mismatch between simulation training and experimental setups can negatively impact the system’s performance. To address this issue, we present a local search method for training optical neural networks in the system. The implementation of this training method allows optical neural networks to reach their state-of-the-art performance levels within the constraints of current experimental settings.

[1]  V. Sorger,et al.  Bistable All‐Optical Devices Based on Nonlinear Epsilon‐Near‐Zero (ENZ) Materials , 2022, Laser & Photonics Reviews.

[2]  Yuhang Li,et al.  Analysis of Diffractive Neural Networks for Seeing Through Random Diffusers , 2022, IEEE Journal of Selected Topics in Quantum Electronics.

[3]  V. Sorger,et al.  Reducing Latency in Sensing for Optical Convolutional Neural Network , 2022, 2022 IEEE Photonics Conference (IPC).

[4]  V. Sorger,et al.  High-density Integrated Photonic Tensor Processing Unit with a Matrix Multiply Compile , 2022, 2022 IEEE Photonics Conference (IPC).

[5]  V. Sorger,et al.  High-performance opto-electronics with emerging materials , 2022, NanoScience + Engineering.

[6]  V. Sorger,et al.  High‐Throughput Multichannel Parallelized Diffraction Convolutional Neural Network Accelerator , 2022, Laser & Photonics Reviews.

[7]  Mario Miscuglio,et al.  Batch processing and data streaming Fourier-based convolutional neural network accelerator , 2022, NanoScience + Engineering.

[8]  B. Bai,et al.  Strategies for training optical neural networks , 2022, National Science Open.

[9]  V. Sorger,et al.  Reconfigurable application-specific photonic integrated circuit for solving partial differential equations , 2022, Nanophotonics.

[10]  V. Sorger,et al.  Self-powered broadband photodetector based on MoS2/Sb2Te3 heterojunctions: a promising approach for highly sensitive detection , 2022, Nanophotonics.

[11]  Xingzhao Liu,et al.  Training optronic convolutional neural networks on an optical system through backpropagation algorithms. , 2022, Optics express.

[12]  R. Tang,et al.  Error-Tolerant Integrated Optical Neural Network Processor based on Multi-Plane Light Conversion , 2022, 2022 Conference on Lasers and Electro-Optics (CLEO).

[13]  V. Sorger,et al.  Waveguide Integrated MoTe2 Photodetector with High speed and energy efficient , 2022, 2022 Conference on Lasers and Electro-Optics (CLEO).

[14]  V. Sorger,et al.  4-bit Ultra Low-loss Cascaded Nonvolatile Photonic Memories Using Broadband Transparent Phase change materials , 2022, 2022 Conference on Lasers and Electro-Optics (CLEO).

[15]  V. Sorger,et al.  Electrical programmable multilevel nonvolatile photonic random-access memory , 2022, Light, science & applications.

[16]  J. Spall,et al.  Hybrid training of optical neural networks , 2022, Optica.

[17]  A. Babakhani,et al.  PIC-based joint transform correlator CNN , 2022, AI and Optical Data Sciences III.

[18]  V. Sorger,et al.  20Gbps high-gain BW-product TMD slot-detector on PIC , 2022, 2D Photonic Materials and Devices V.

[19]  Ray T. Chen,et al.  Design and experimental demonstration of a hardware-efficient integrated optical neural network , 2022, Smart Photonic and Optoelectronic Integrated Circuits 2022.

[20]  V. Sorger,et al.  Photonic Tensor Core with Photonic Compute-in-Memory , 2022, 2022 Optical Fiber Communications Conference and Exhibition (OFC).

[21]  V. Sorger,et al.  Highly accurate, reliable, and non-contaminating two-dimensional material transfer system , 2022, Applied Physics Reviews.

[22]  Volker J. Sorger,et al.  Emerging Devices and Packaging Strategies for Electronic-Photonic AI Accelerators , 2021, ArXiv.

[23]  V. Sorger,et al.  100 GHz micrometer-compact broadband monolithic ITO Mach–Zehnder interferometer modulator enabling 3500 times higher packing density , 2021, 2112.10926.

[24]  T. Shimobaba,et al.  Hyperparameter tuning of optical neural network classifiers for high-order Gaussian beams. , 2021, Optics express.

[25]  V. Sorger,et al.  Self-driven highly responsive PN junction InSe heterostructure near-infrared light detector , 2021, 2D Photonic Materials and Devices V.

[26]  Mitsuru Takenaka,et al.  Si microring resonator crossbar array for on-chip inference and training of optical neural network , 2021, ACS Photonics.

[27]  Logan G. Wright,et al.  An optical neural network using less than 1 photon per multiplication , 2021, Nature Communications.

[28]  V. Sorger,et al.  Training on System for Opto-Electrical Neural Network , 2022, Imaging and Applied Optics Congress 2022 (3D, AOA, COSI, ISA, pcAOP).

[29]  V. Sorger,et al.  Design and Testing of Integrated Photonic Chip for Convolution Neural Network , 2022, Imaging and Applied Optics Congress 2022 (3D, AOA, COSI, ISA, pcAOP).

[30]  V. Sorger,et al.  Electrical Programmable Low-loss high cyclable Nonvolatile Photonic Random-Access Memory , 2022 .

[31]  V. Sorger,et al.  Electrical Pulse Driven Multi-Level Nonvolatile Photonic Memories Using Broadband Transparent Phase Change Materials , 2022 .

[32]  V. Sorger,et al.  An ITO - Graphene heterojunction integrated absorption modulator on Si-photonics for neuromorphic nonlinear activation , 2021, APL Photonics.

[33]  Kostas Daniilidis,et al.  Self-Supervised Optical Flow with Spiking Neural Networks and Event Based Cameras , 2021, 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[34]  V. Sorger,et al.  Compact Plasmonic Enhanced MoTe2 Photodetector based on Engineering Gain-Bandwidth-Product Scaling Laws , 2021, 2021 26th Microoptics Conference (MOC).

[35]  V. Sorger,et al.  1fJ/bit Coupling-based ITO Monolithic Modulator in Integrated Photonics , 2021, 2021 26th Microoptics Conference (MOC).

[36]  V. Sorger,et al.  Highly Accurate, Reliable and Non-Contaminating Two-Dimensional Material Transfer System , 2021, 2021 26th Microoptics Conference (MOC).

[37]  V. Sorger,et al.  Strain-Induced Spatially Resolved Charge Transport in 2H-MoTe2 , 2021, ACS Applied Electronic Materials.

[38]  T. Itoh,et al.  Approximate analog computing with metatronic circuits , 2021, Communications Physics.

[39]  Aydogan Ozcan,et al.  Design of Scale-, Shift- and Rotation-Invariant Diffractive Optical Networks , 2021, Emerging Topics in Artificial Intelligence (ETAI) 2021.

[40]  Mario Miscuglio,et al.  Prospects and applications of photonic neural networks , 2021, Advances in Physics: X.

[41]  V. Sorger,et al.  Effect of Strain in WS2 Monolayer Integrated Excitonic Photodetector , 2021, 2021 Conference on Lasers and Electro-Optics (CLEO).

[42]  V. Sorger,et al.  Performance Analysis of Integrated Electro-Optic Phase Modulators Based on Emerging Materials , 2021, IEEE Journal of Selected Topics in Quantum Electronics.

[43]  David Moss,et al.  Advanced applications of microcombs: From optical neural networks to data transmission , 2021 .

[44]  Hong Cai,et al.  Efficient On-Chip Training of Optical Neural Networks Using Genetic Algorithm , 2021, ACS Photonics.

[45]  Ray T. Chen,et al.  Towards lab-on-chip ultrasensitive ethanol detection using photonic crystal waveguide operating in the mid-infrared , 2021, Nanophotonics.

[46]  Zejun Zhang,et al.  Advances and marine applications of optical neural network , 2021, Symposium on Novel Photoelectronic Detection Technology and Application.

[47]  V. Sorger,et al.  Symmetry perception with spiking neural networks , 2021, Scientific Reports.

[48]  Jun Luo,et al.  Robust light beam diffractive shaping based on a kind of compact all-optical neural network , 2021, Optics Express.

[49]  G. Lo,et al.  An optical neural chip for implementing complex-valued neural network , 2021, Nature Communications.

[50]  Ray T. Chen,et al.  Efficient On-Chip Learning for Optical Neural Networks Through Power-Aware Sparse Zeroth-Order Optimization , 2020, AAAI.

[51]  Ray T. Chen,et al.  Heterogeneously integrated ITO plasmonic Mach–Zehnder interferometric modulator on SOI , 2020, Scientific reports.

[52]  A. Babakhani,et al.  Electronic Bottleneck Suppression in Next‐Generation Networks with Integrated Photonic Digital‐to‐Analog Converters , 2019, Advanced Photonics Research.

[53]  V. Sorger,et al.  Complex Exponential Neural Network for Optical System , 2021, Photonics in Switching and Computing 2021.

[54]  V. Sorger,et al.  Channel Tiling for Improved Performance and Accuracy of Optical Neural Network Accelerators , 2020, ArXiv.

[55]  V. Sorger,et al.  GHz-Fast Multiplexed OAM Generation and Detection , 2020 .

[56]  Mario Miscuglio,et al.  Electro-Optical Hybrid Fourier Neural Network with Amplitude-Only Modulation , 2020 .

[57]  V. Sorger,et al.  Efficient MoTe2 Slot-enhanced Photodetector based on Engineering Gain-Bandwidth-Product Scaling Laws , 2020 .

[58]  V. Sorger,et al.  Massively parallel amplitude-only Fourier neural network , 2020, AI and Optical Data Sciences II.

[59]  F. Horst,et al.  Opportunities for integrated photonic neural networks , 2020 .

[60]  Ray T. Chen,et al.  Broadband Sub-λ GHz ITO Plasmonic Mach-Zehnder Modulator in Silicon Photonics , 2020 .

[61]  Tarek El-Ghazawi,et al.  Analog Computing with Metatronic Circuits , 2020, ArXiv.

[62]  Jin Zhang,et al.  Inverse design of an integrated-nanophotonics optical neural network. , 2020, Science bulletin.

[63]  Tarek El-Ghazawi,et al.  A Design Methodology for Post-Moore’s Law Accelerators: The Case of a Photonic Neuromorphic Processor , 2020, 2020 IEEE 31st International Conference on Application-specific Systems, Architectures and Processors (ASAP).

[64]  Tao Yan,et al.  In situ optical backpropagation training of diffractive optical neural networks , 2020 .

[65]  V. Sorger,et al.  2D Material Based Electro-Absorption Modulator in Si Photonics , 2020, 2020 Conference on Lasers and Electro-Optics (CLEO).

[66]  V. Sorger,et al.  WS2 Monolayer Integrated Photodetector , 2020, 2020 Conference on Lasers and Electro-Optics (CLEO).

[67]  Mario Miscuglio,et al.  GSST-Based Photonic Memory Multilevel Perceptron , 2020, 2020 Conference on Lasers and Electro-Optics (CLEO).

[68]  Puneet Gupta,et al.  Million-Channel Parallelism Fourier-Optic Convolutional Filter and Neural Network Processor , 2020, 2020 Conference on Lasers and Electro-Optics (CLEO).

[69]  Dirk Englund,et al.  Digital Optical Neural Networks for Large-Scale Machine Learning , 2020, 2020 Conference on Lasers and Electro-Optics (CLEO).

[70]  A. Boes,et al.  Photonic perceptron based on a Kerr microcomb for high-speed, scalable, optical neural networks , 2020, 2020 International Topical Meeting on Microwave Photonics (MWP).

[71]  Jiaqi Gu,et al.  ROQ: A Noise-Aware Quantization Scheme Towards Robust Optical Neural Networks with Low-bit Controls , 2020, 2020 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[72]  S. Solares,et al.  Strain-engineered high-responsivity MoTe2 photodetector for silicon photonic integrated circuits , 2019, Nature Photonics.

[73]  Yifei Zhang,et al.  Intelligent edge processing with photonic multilevel memory , 2020 .

[74]  Soon Thor Lim,et al.  Convolutional Neural Networks With Integrated Optics , 2020 .

[75]  A. Lvovsky,et al.  Backpropagation through nonlinear units for the all-optical training of neural networks , 2019, Photonics Research.

[76]  Chii-Chang Chen,et al.  Optical neural networks based on optical fiber-communication system , 2019, Neurocomputing.

[77]  Sasikanth Manipatruni,et al.  Design of optical neural networks with component imprecisions , 2019, Optics express.

[78]  Xu Wang,et al.  Self-learning photonic signal processor with an optical neural network chip , 2019, 1902.07318.

[79]  Shanhui Fan,et al.  Training of Photonic Neural Networks through In Situ Backpropagation , 2018, 2019 Conference on Lasers and Electro-Optics (CLEO).

[80]  Elliot Meyerson,et al.  Evolving Deep Neural Networks , 2017, Artificial Intelligence in the Age of Neural Networks and Brain Computing.

[81]  Mario Miscuglio,et al.  All-optical nonlinear activation function for photonic neural networks [Invited] , 2018, Optical Materials Express.

[82]  Zhenming Yu,et al.  An integrated optical neural network chip based on Mach-Zehnder interferometers , 2018, 2018 Asia Communications and Photonics Conference (ACP).

[83]  W. Heidrich,et al.  Hybrid optical-electronic convolutional neural networks with optimized diffractive optics for image classification , 2018, Scientific Reports.

[84]  Yi Luo,et al.  All-optical machine learning using diffractive deep neural networks , 2018, Science.

[85]  Tareq Abed Mohammed,et al.  Understanding of a convolutional neural network , 2017, 2017 International Conference on Engineering and Technology (ICET).

[86]  Yurong Liu,et al.  A survey of deep neural network architectures and their applications , 2017, Neurocomputing.

[87]  Dirk Englund,et al.  Deep learning with coherent nanophotonic circuits , 2017, 2017 Fifth Berkeley Symposium on Energy Efficient Electronic Systems & Steep Transistors Workshop (E3S).

[88]  Eugenio Culurciello,et al.  An Analysis of Deep Neural Network Models for Practical Applications , 2016, ArXiv.

[89]  Joni Dambre,et al.  Towards integrated optical spiking neural networks: Delaying spikes on chip , 2013, 2013 Sixth "Rio De La Plata" Workshop on Laser Dynamics and Nonlinear Photonics.

[90]  Yoshua Bengio,et al.  Exploring Strategies for Training Deep Neural Networks , 2009, J. Mach. Learn. Res..

[91]  Steven R. Skinner,et al.  Reinforcement and backpropagation training for an optical neural network using self-lensing effects , 2000, IEEE Trans. Neural Networks Learn. Syst..

[92]  Silvano Martello,et al.  Meta-Heuristics: Advances and Trends in Local Search Paradigms for Optimization , 2012 .

[93]  Yves Crama,et al.  Local Search in Combinatorial Optimization , 2018, Artificial Neural Networks.

[94]  W. Hart.  Adaptive global optimization with local search , 1994 .

[95]  Charles M. Bachmann,et al.  Neural Networks and Their Applications , 1994 .

[96]  D Psaltis,et al.  Local learning algorithm for optical neural networks. , 1992, Applied optics.

[97]  Robert Hecht-Nielsen,et al.  Theory of the backpropagation neural network , 1989, International 1989 Joint Conference on Neural Networks.

[98]  Jehoshua Bruck,et al.  A study on neural networks , 1988, Int. J. Intell. Syst..