Class-specific Differential Detection in Diffractive Optical Neural Networks Improves Inference Accuracy

Abstract

Diffractive deep neural networks have been introduced earlier as an optical machine learning framework that uses task-specific diffractive surfaces designed by deep learning to all-optically perform inference, achieving promising performance for object classification and imaging. Here we demonstrate systematic improvements in diffractive optical neural networks based on a differential measurement technique that mitigates the non-negativity constraint of light intensity. In this scheme, each class is assigned to a separate pair of photodetectors, behind a diffractive network, and the class inference is made by maximizing the normalized signal difference between the detector pairs. Moreover, by utilizing the inherent parallelization capability of optical systems, we reduced the signal coupling between the positive and negative detectors of each class by dividing their optical path into two jointly-trained diffractive neural networks that work in parallel. We further made use of this parallelization approach, and divided individual classes among multiple jointly-trained differential diffractive neural networks. Using this class-specific differential detection in jointly-optimized diffractive networks, our simulations achieved testing accuracies of 98.52%, 91.48% and 50.82% for MNIST, Fashion-MNIST and grayscale CIFAR-10 datasets, respectively. Similar to ensemble methods practiced in machine learning, we also independently-optimized multiple differential diffractive networks that optically project their light onto a common detector plane, and achieved testing accuracies of 98.59%, 91.06% and 51.44% for MNIST, Fashion-MNIST and grayscale CIFAR-10, respectively. Through these systematic advances in designing diffractive neural networks, the reported classification accuracies set the state-of-the-art for an all-optical neural network design.

21 pages, 6 Figures, 3 Tables

References in corpus (12)

• Adam: A Method for Stochastic Optimization
• A Survey on Deep Learning in Medical Image Analysis
• Distilling the Knowledge in a Neural Network
• Deep Learning with Coherent Nanophotonic Circuits
• All-Optical Machine Learning Using Diffractive Deep Neural Networks
• Neuromorphic Silicon Photonic Networks
• Training of photonic neural networks through in situ backpropagation
• Reinforcement Learning in a large scale photonic Recurrent Neural Network
• Wave Physics as an Analog Recurrent Neural Network
• Analysis of Diffractive Optical Neural Networks and Their Integration with Electronic Neural Networks
• Toward Fast Neural Computing using All-Photonic Phase Change Spiking Neurons
• On-Chip Optical Convolutional Neural Networks

Cited by in corpus (26)

• 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
• 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
• 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
• Universal Linear Intensity Transformations Using Spatially-Incoherent Diffractive Processors
• Multiple-image encryption and hiding with an optical diffractive neural network
• Cascadable all-optical NAND gates using diffractive networks
• Classification and reconstruction of spatially overlapping phase images using diffractive optical networks
• Information hiding cameras: optical concealment of object information into ordinary images
• Simultaneously sorting vector vortex beams of 120 modes
• Ultrahigh-fidelity spatial mode quantum gates in high-dimensional space by diffractive deep neural networks
• Pyramid diffractive optical networks for unidirectional image magnification and demagnification
• Complex-valued universal linear transformations and image encryption using spatially incoherent diffractive networks
• All-Optical Phase Conjugation Using Diffractive Wavefront Processing
• All-Optical Image Identification with Programmable Matrix Transformation
• Express Wavenet -- a low parameter optical neural network with random shift wavelet pattern
• K-series approximation of vectorial optical fields for designing diffractive optical elements with subwavelength feature sizes
• A Diffractive Neural Network with Weight-Noise-Injection Training
• Phase-multiplexed optical computing: Reconfiguring a multi-task diffractive optical processor using illumination phase diversity