Toward Fast Neural Computing using All-Photonic Phase Change Spiking Neurons

TL;DR

This paper introduces a novel all-photonic spiking neuron using phase change materials, enabling ultrafast neural computing with high bandwidth and energy efficiency, advancing photonic neuromorphic hardware.

Contribution

It demonstrates a purely photonic Integrate-and-Fire neuron based on phase change dynamics, enabling fast, energy-efficient neuromorphic computing in the photonic domain.

Findings

Photonic neuron operates based on phase change dynamics of GST.

Potential for integration into all-photonic neural networks.

Promises ultrafast inference with high bandwidth.

Abstract

The rapid growth of brain-inspired computing coupled with the inefficiencies in the CMOS implementations of neuromrphic systems has led to intense exploration of efficient hardware implementations of the functional units of the brain, namely, neurons and synapses. However, efforts have largely been invested in implementations in the electrical domain with potential limitations of switching speed, packing density of large integrated systems and interconnect losses. As an alternative, neuromorphic engineering in the photonic domain has recently gained attention. In this work, we demonstrate a purely photonic operation of an Integrate-and-Fire Spiking neuron, based on the phase change dynamics of Ge$_2$Sb$_2$Te$_5$ (GST) embedded on top of a microring resonator, which alleviates the energy constraints of PCMs in electrical domain. We also show that such a neuron can be potentially integrated with on-chip synapses into an all-Photonic Spiking Neural network inferencing framework which promises to be ultrafast and can potentially offer a large operating bandwidth.