Analysis of Diffractive Optical Neural Networks and Their Integration with Electronic Neural Networks

Abstract

Optical machine learning offers advantages in terms of power efficiency, scalability and computation speed. Recently, an optical machine learning method based on Diffractive Deep Neural Networks (D2NNs) has been introduced to execute a function as the input light diffracts through passive layers, designed by deep learning using a computer. Here we introduce improvements to D2NNs by changing the training loss function and reducing the impact of vanishing gradients in the error back-propagation step. Using five phase-only diffractive layers, we numerically achieved a classification accuracy of 97.18% and 89.13% for optical recognition of handwritten digits and fashion products, respectively; using both phase and amplitude modulation (complex-valued) at each layer, our inference performance improved to 97.81% and 89.32%, respectively. Furthermore, we report the integration of D2NNs with electronic neural networks to create hybrid-classifiers that significantly reduce the number of input pixels into an electronic network using an ultra-compact front-end D2NN with a layer-to-layer distance of a few wavelengths, also reducing the complexity of the successive electronic network. Using a 5-layer phase-only D2NN jointly-optimized with a single fully-connected electronic layer, we achieved a classification accuracy of 98.71% and 90.04% for the recognition of handwritten digits and fashion products, respectively. Moreover, the input to the electronic network was compressed by >7.8 times down to 10x10 pixels. Beyond creating low-power and high-frame rate machine learning platforms, D2NN-based hybrid neural networks will find applications in smart optical imager and sensor design.

arXiv comment22 pages, 5 Figures, 4 Tables, 1 Supplementary Figure, 2 Supplementary Tables

References in corpus (15)

• Adam: A Method for Stochastic Optimization
• Deep Learning in Neural Networks: An Overview
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Deep Learning with Coherent Nanophotonic Circuits
• All-Optical Machine Learning Using Diffractive Deep Neural Networks
• Neuromorphic Silicon Photonic Networks
• Deep Learning using Linear Support Vector Machines
• Training of photonic neural networks through in situ backpropagation
• Reinforcement Learning in a large scale photonic Recurrent Neural Network
• Programmable unitary spatial modes manipulation
• Complex wavefront engineering with disorder-engineered metasurfaces
• Toward Fast Neural Computing using All-Photonic Phase Change Spiking Neurons
• PCNNA: A Photonic Convolutional Neural Network Accelerator
• On-Chip Optical Convolutional Neural Networks
• Principles of Neuromorphic Photonics

Cited by in corpus (48)

• Design of Task-Specific Optical Systems Using Broadband Diffractive Neural Networks
• Spectrally-Encoded Single-Pixel Machine Vision Using Diffractive Networks
• Terahertz Pulse Shaping Using Diffractive Surfaces
• Computational Imaging Without a Computer: Seeing Through Random Diffusers at the Speed of Light
• Class-specific Differential Detection in Diffractive Optical Neural Networks Improves Inference Accuracy
• All-Optical Information Processing Capacity of Diffractive Surfaces
• Misalignment Resilient Diffractive Optical Networks
• Ensemble learning of diffractive optical networks
• Polarization Multiplexed Diffractive Computing: All-Optical Implementation of a Group of Linear Transformations Through a Polarization-Encoded Diffractive Network
• Snapshot Multispectral Imaging Using a Diffractive Optical Network
• Diffractive all-optical computing for quantitative phase imaging
• Massively Parallel Universal Linear Transformations using a Wavelength-Multiplexed Diffractive Optical Network
• Scale-, shift- and rotation-invariant diffractive optical networks
• Rapid Sensing of Hidden Objects and Defects using a Single-Pixel Diffractive Terahertz Processor
• All-optical image classification through unknown random diffusers using a single-pixel diffractive network
• Universal Linear Intensity Transformations Using Spatially-Incoherent Diffractive Processors
• Super-resolution image display using diffractive decoders
• All-optical image denoising using a diffractive visual processor
• Data class-specific all-optical transformations and encryption
• Unidirectional Imaging using Deep Learning-Designed Materials
• Multiplane Quantitative Phase Imaging Using a Wavelength-Multiplexed Diffractive Optical Processor
• Cascadable all-optical NAND gates using diffractive networks
• Optical Generative Models
• Photonic convolutional neural networks using integrated diffractive optics
• Universal Polarization Transformations: Spatial programming of polarization scattering matrices using a deep learning-designed diffractive polarization transformer
• Classification and reconstruction of spatially overlapping phase images using diffractive optical networks
• Quantitative phase imaging (QPI) through random diffusers using a diffractive optical network
• Subwavelength Imaging using a Solid-Immersion Diffractive Optical Processor
• Pyramid diffractive optical networks for unidirectional image magnification and demagnification
• A Broad-Spectrum Diffractive Network via Ensemble Learning
• Broadband nonlinear modulation of incoherent light using a transparent optoelectronic neuron array
• Time-lapse image classification using a diffractive neural network
• Multiplexed all-optical permutation operations using a reconfigurable diffractive optical network
• An optoacoustic field-programmable perceptron for recurrent neural networks
• Analysis of Diffractive Neural Networks for Seeing Through Random Diffusers
• Integration of Programmable Diffraction with Digital Neural Networks
• Design of ultracompact broadband focusing spectrometers based on deep diffractive neural networks
• Metameric Varifocal Holography
• Express Wavenet -- a low parameter optical neural network with random shift wavelet pattern
• Unidirectional imaging with partially coherent light
• An optical diffractive deep neural network with multiple frequency-channels
• Broadband Unidirectional Visible Imaging Using Wafer-Scale Nano-Fabrication of Multi-Layer Diffractive Optical Processors
• Coherence Awareness in Diffractive Neural Networks
• Computer-free, all-optical reconstruction of holograms using diffractive networks
• Snapshot 3D image projection using a diffractive decoder
• High speed and reconfigurable optronic neural network with digital nonlinear activation
• Joint Image and Depth Estimation with Mask-Based Lensless Cameras
• Numerical study of laser micro- and nano-processing of nanocomposite porous materials