A spiking photonic neural network of 40.000 neurons, trained with rank-order coding for leveraging sparsity

Tools

Talukder, Ria and Skalli, Anas and Porte, Xavier and Thorpe, Simon and Brunner, Daniel (2024) A spiking photonic neural network of 40.000 neurons, trained with rank-order coding for leveraging sparsity. Other. arXiv, Ithaca, NY. (https://doi.org/10.48550/arXiv.2411.19209)

[thumbnail of Talukder-etal-ArXiv-2024-A-spiking-photonic-neural-network-of-40000-neurons]
Preview
 Text. Filename: Talukder-etal-ArXiv-2024-A-spiking-photonic-neural-network-of-40000-neurons.pdf
Final Published Version
License: Creative Commons Attribution 4.0 logo
Download (1MB)| Preview

Abstract

In recent years, the hardware implementation of neural networks, leveraging physical coupling and analog neurons has substantially increased in relevance. Such nonlinear and complex physical networks provide significant advantages in speed and energy efficiency, but are potentially susceptible to internal noise when compared to digital emulations of such networks. In this work, we consider how additive and multiplicative Gaussian white noise on the neuronal level can affect the accuracy of the network when applied for specific tasks and including a softmax function in the readout layer. We adapt several noise reduction techniques to the essential setting of classification tasks, which represent a large fraction of neural network computing. We find that these adjusted concepts are highly effective in mitigating the detrimental impact of noise.

ORCID iDs

Talukder, Ria, Skalli, Anas, Porte, Xavier ORCID logoORCID: https://orcid.org/0000-0002-9869-7170, Thorpe, Simon and Brunner, Daniel;
CORE (COnnecting REpositories)