Unsupervised Image Classification Through Time-Multiplexed Photonic Multi-Layer Spiking Convolutional Neural Network

TL;DR

This paper demonstrates an unsupervised deep photonic spiking CNN using VCSELs, achieving significant reduction in physical neuron count through time-multiplexing and fast response, for image classification.

Contribution

It introduces a novel photonic spiking CNN architecture with unsupervised learning and neuron time-multiplexing, reducing physical neuron count by 90%.

Findings

Successful implementation of a deep photonic spiking CNN.

Achieved 90% reduction in physical neuron count.

Demonstrated effective unsupervised image classification.

Abstract

We present results of a deep photonic spiking convolutional neural network, based on two-section VCSELs, targeting image classification. Training is based on unsupervised spike-timing dependent plasticity, whereas neuron time-multiplexing and ultra-fast response are exploited towards a a reduction of the physical neuron count by 90%