On-Chip Optical Convolutional Neural Networks

Abstract

Convolutional Neural Networks (CNNs) are a class of Artificial Neural Networks(ANNs) that employ the method of convolving input images with filter-kernels for object recognition and classification purposes. In this paper, we propose a photonics circuit architecture which could consume a fraction of energy per inference compared with state of the art electronics.

arXiv comment18 pages, 7 figures

References in corpus (1)

• Quantum transport simulations in a programmable nanophotonic processor

Cited by in corpus (12)

• Ultrafast and energy-efficient all-optical switching with graphene-loaded deep-subwavelength plasmonic waveguides
• Large-Scale Optical Neural Networks based on Photoelectric Multiplication
• Analysis of Diffractive Optical Neural Networks and Their Integration with Electronic Neural Networks
• Digital Electronics and Analog Photonics for Convolutional Neural Networks (DEAP-CNNs)
• Class-specific Differential Detection in Diffractive Optical Neural Networks Improves Inference Accuracy
• Photonic convolutional neural networks using integrated diffractive optics
• Photonic Convolution Neural Network Based on Interleaved Time-Wavelength Modulation
• Backpropagation through nonlinear units for all-optical training of neural networks
• Channel Tiling for Improved Performance and Accuracy of Optical Neural Network Accelerators
• All-optical neuromorphic binary convolution with a spiking VCSEL neuron for image gradient magnitudes
• Performance evaluation of an integrated photonic convolutional neural network based on delay buffering and wavelength division multiplexing
• End-to-End Framework for Efficient Deep Learning Using Metasurfaces Optics