Photonic Perceptron Based on a Kerr Microcomb for High-Speed, Scalable, Optical Neural Networks

摘要

Optical artificial neural networks (ONNs)-analog computing hardware tailored for machine learning-have significant potential for achieving ultrahigh computing speed and energy efficiency. A new approach to architectures for ONNs based on integrated Kerr microcomb sources that is programmable, highly scalable, and capable of reaching ultra-high speeds is proposed here. The building block of the ONN-a single neuron perceptron-is experimentally demonstrated that reaches a high single-unit throughput speed of 11.9 Giga-FLOPS at 8 bits per FLOP, corresponding to 95.2 Gbps, achieved by mapping synapses onto 49 wavelengths of a microcomb. The perceptron is tested on simple standard benchmark datasets-handwritten-digit recognition and cancer-cell detection-achieving over 90% and 85% accuracy, respectively. This performance is a direct result of the record low wavelength spacing (49 GHz) for a coherent integrated microcomb source, which results in an unprecedented number of wavelengths for neuromorphic optics. Finally, an approach to scaling the perceptron to a deep learning network is proposed using the same single microcomb device and standard off-the-shelf telecommunications technology, for high-throughput operation involving full matrix multiplication for applications such as real-time massive data processing for unmanned vehicles and aircraft tracking.