Electrically programmable phase-change photonic memory for optical neural networks with nanoseconds in situ training capability
Advanced Photonics · 18.07.2023 · 5 (04) · 1-9
DOI: 10.1117/1.ap.5.4.046004 ↗Kurz gefasst
Fachartikel in der Zeitschrift Advanced Photonics, erschienen am 18.07.2023. Verfasst von Maoliang Wei und 20 weiteren Autor:innen. Die Arbeit wurde bisher 109-mal zitiert (Crossref) und nennt 59 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Optical neural networks (ONNs), enabling low latency and high parallel data processing without electromagnetic interference, have become a viable player for fast and energy-efficient processing and calculation to meet the increasing demand for hash rate. Photonic memories employing nonvolatile phase-change materials could achieve zero static power consumption, low thermal cross talk, large-scale, and high-energy-efficient photonic neural networks. Nevertheless, the switching speed and dynamic energy consumption of phase-change material-based photonic memories make them inapplicable for in situ training. Here, by integrating a patch of phase change thin film with a PIN-diode-embedded microring resonator, a bifunctional photonic memory enabling both 5-bit storage and nanoseconds volatile modulation was demonstrated. For the first time, a concept is presented for electrically programmable phase-change material-driven photonic memory integrated with nanosecond modulation to allow fast in situ training and zero static power consumption data processing in ONNs. ONNs with an optical convolution kernel constructed by our photonic memory theoretically achieved an accuracy of predictions higher than 95% when tested by the MNIST handwritten digit database. This provides a feasible solution to constructing large-scale nonvolatile ONNs with high-speed in situ training capability.
Zitieren
APA
Wei, M., Li, J., Chen, Z., Tang, B., Jia, Z., Zhang, P., Lei, K., Xu, K., Wu, J., Zhong, C., Ma, H., Ye, Y., Jian, J., Sun, C., Liu, R., Sun, Y., Sha, W. E. I., Hu, X., Yang, J., …, & Lin, H. (2023). Electrically programmable phase-change photonic memory for optical neural networks with nanoseconds in situ training capability. Advanced Photonics, 5(04), 1-9. https://doi.org/10.1117/1.ap.5.4.046004
Vancouver
Wei M, Li J, Chen Z, Tang B, Jia Z, Zhang P, et al. Electrically programmable phase-change photonic memory for optical neural networks with nanoseconds in situ training capability. Advanced Photonics. 2023;5(04):1-9. doi:10.1117/1.ap.5.4.046004
BibTeX
@article{Wei2023,
author = {Wei, Maoliang and Li, Junying and Chen, Zequn and Tang, Bo and Jia, Zhiqi and Zhang, Peng and Lei, Kunhao and Xu, Kai and Wu, Jianghong and Zhong, Chuyu and Ma, Hui and Ye, Yuting and Jian, Jialing and Sun, Chunlei and Liu, Ruonan and Sun, Ying and Sha, Wei. E. I. and Hu, Xiaoyong and Yang, Jianyi and Li, Lan and Lin, Hongtao},
title = {Electrically programmable phase-change photonic memory for optical neural networks with nanoseconds in situ training capability},
journal = {Advanced Photonics},
year = {2023},
volume = {5},
number = {04},
pages = {1--9},
doi = {10.1117/1.ap.5.4.046004}
}
EUVIMED