Stochastic photonic spiking neuron for Bayesian inference with unsupervised learning
Optics Letters · 2023
DOI: 10.1364/ol.484268 ↗Kurz gefasst
Fachartikel in der Zeitschrift Optics Letters, 2023. Verfasst von Bowen Ma und 3 weiteren Autor:innen. Hauptthemen laut MEDLINE: Unsupervised Machine Learning, Neurons. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 17 Literaturangaben. Gefördert von: National Key Research and Development Program of China, National Natural Science Foundation of China.
Abstract (Originalsprache)
Stochasticity is an inherent feature of biological neural activities. We propose a noise-injection scheme to implement a GHz-rate stochastic photonic spiking neuron (S-PSN). The firing-probability encoding is experimentally demonstrated and exploited for Bayesian inference with unsupervised learning. In a breast diagnosis task, the stochastic photonic spiking neural network (S-PSNN) can not only achieve a classification accuracy of 96.6%, but can also evaluate the diagnosis uncertainty with prediction entropies. As a result, the misdiagnosis rate is reduced by 80% compared to that of a conventional deterministic photonic spiking neural network (D-PSNN) for the same task. The GHz-rate S-PSN endows the neuromorphic photonics with high-speed Bayesian inference for reliable information processing in error-critical scenarios.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Förderung
- National Key Research and Development Program of China
- National Natural Science Foundation of China
Zitieren
APA
Ma, B., Zhang, J., Li, X., & Zou, W. (2023). Stochastic photonic spiking neuron for Bayesian inference with unsupervised learning. Optics Letters. https://doi.org/10.1364/ol.484268
Vancouver
Ma B, Zhang J, Li X, Zou W. Stochastic photonic spiking neuron for Bayesian inference with unsupervised learning. Optics Letters. 2023. doi:10.1364/ol.484268
BibTeX
@article{Ma2023,
author = {Ma, Bowen and Zhang, Junfeng and Li, Xing and Zou, Weiwen},
title = {Stochastic photonic spiking neuron for Bayesian inference with unsupervised learning},
journal = {Optics Letters},
year = {2023},
doi = {10.1364/ol.484268}
}
EUVIMED