Magnetization reversal by multiple optical pulses for photonic spiking neuron with the leaky integrate and fire model DOI

Gaku Takagi,

Toshiya Murai, Yuya Shoji

et al.

Japanese Journal of Applied Physics, Journal Year: 2024, Volume and Issue: 63(7), P. 072002 - 072002

Published: June 3, 2024

Abstract Photonic accelerators are anticipated to be the next generation of hardware processors, replacing traditional electronic accelerators. In current photonic based on artificial neural networks, integrated circuits incorporated with leverage their strengths: used perform linear calculations, while nonlinear calculations. However, this architecture requires optoelectric conversion at each layer and is unable superiority light. We propose a novel spiking neuron magneto-optical synapse an all-optical network. This study experimentally demonstrates that magnetization reversal CoFeB, which occurs during thermal accumulation owing multiple optical pulses, similar behavior leaky fire model neurons.

Language: Английский

Study on the aging status of insulators based on hyperspectral imaging technology DOI Creative Commons
Yihan Fan, Yujun Guo, Yang Liu

et al.

Optics Express, Journal Year: 2024, Volume and Issue: 32(4), P. 5072 - 5072

Published: Jan. 22, 2024

The acidic environment is one of the main factors leading to aging silicone rubber (SiR) insulators. Aging can reduce surface hydrophobicity and pollution flashover resistance insulators, threatening safe stable operation power grid. Therefore, evaluating state insulators essential prevent accidents on transmission line. This paper based an optical hyperspectral imaging (HSI) technology for pixel-level assessment insulator status. Firstly, SiR samples were artificially aged in three typical solutions with different concentrations HNO 3 , H 2 SO 4 HCl, six grades prepared. HSI at each grade was extracted using a imager. To calculation complexity eliminate interference useless information band, this proposes joint random forest- principal component analysis (RF-PCA) dimensionality reduction method original 256-dimensional data 7 dimensions. Finally, capture local features images more effectively retain most significant spectral lines, convolutional neural network (CNN) used build classification model SiR's visual prediction insulators’ defects. research provides important guarantee timely detection safety hazards

Language: Английский

Citations

4
Magnetization reversal by multiple optical pulses for photonic spiking neuron with the leaky integrate and fire model DOI

Gaku Takagi,

Toshiya Murai, Yuya Shoji

et al.

Japanese Journal of Applied Physics, Journal Year: 2024, Volume and Issue: 63(7), P. 072002 - 072002

Published: June 3, 2024

Abstract Photonic accelerators are anticipated to be the next generation of hardware processors, replacing traditional electronic accelerators. In current photonic based on artificial neural networks, integrated circuits incorporated with leverage their strengths: used perform linear calculations, while nonlinear calculations. However, this architecture requires optoelectric conversion at each layer and is unable superiority light. We propose a novel spiking neuron magneto-optical synapse an all-optical network. This study experimentally demonstrates that magnetization reversal CoFeB, which occurs during thermal accumulation owing multiple optical pulses, similar behavior leaky fire model neurons.

Language: Английский

Citations

1