Photonic convolutional neural networks using integrated diffractive optics

Abstract

With recent rapid advances in photonic integrated circuits, it has been demonstrated that programmable photonic chips can be used to implement artificial neural networks. Convolutional neural networks (CNN) are a class of deep learning methods that have been highly successful in applications such as image classification and speech processing. We present an architecture to implement a photonic CNN using the Fourier transform property of integrated star couplers. We show, in computer simulation, high accuracy image classification using the MNIST dataset. We also model component imperfections in photonic CNN and show that the performance degradation can be recovered in a programmable chip. Our proposed architecture provides a large reduction in physical footprint compared to current implementations as it utilizes the natural advantages of optics and hence offers a scalable pathway towards integrated photonic deep learning processors.

9 pages, 6 figures

References in corpus (25)

• Adam: A Method for Stochastic Optimization
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• TensorFlow: A system for large-scale machine learning
• 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
• Universal Linear Optics
• Neuromorphic Silicon Photonic Networks
• Convolutional Networks for Fast, Energy-Efficient Neuromorphic Computing
• Training of photonic neural networks through in situ backpropagation
• Self-Configuring Universal Linear Optical Component
• Reprogrammable Electro-Optic Nonlinear Activation Functions for Optical Neural Networks
• A silicon photonic modulator neuron
• Electric Field-Induced Second Order Nonlinear Optical Effects in Silicon Waveguides
• Analysis of Diffractive Optical Neural Networks and Their Integration with Electronic Neural Networks
• All-optical mode unscrambling on a silicon photonic chip
• Unitary Evolution Recurrent Neural Networks
• Spectral Representations for Convolutional Neural Networks
• Design of optical neural networks with component imprecisions
• Random Projections through multiple optical scattering: Approximating kernels at the speed of light
• An Optical Frontend for a Convolutional Neural Network
• On-Chip Optical Convolutional Neural Networks
• Benchmarking integrated photonic architectures
• Analog Optical Computing Primitives in Silicon Photonics
• Parallel fault-tolerant programming of an arbitrary feedforward photonic network

Cited by in corpus (5)

• Ensemble learning of diffractive optical networks
• Classification and reconstruction of spatially overlapping phase images using diffractive optical networks
• Self-learning Machines based on Hamiltonian Echo Backpropagation
• Learning Arbitrary Complex Matrices by Interlacing Amplitude and Phase Masks with Fixed Unitary Operations
• Planar Fourier Optics for Slab Waveguides, Surface Plasmon Polaritons and 2D Materials