EUVIMEDEUVIMED entdecken
Crossref

Hybrid neuromorphic hardware with sparing 2D synapse and CMOS neuron for character recognition

Siwei Xue, Shuiyuan Wang, Tianxiang Wu, Ziye Di, Nuo Xu, Yibo Sun, Chaofan Zeng, Shunli Ma, Peng Zhou

Science Bulletin · 2023

DOI: 10.1016/j.scib.2023.09.006 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science Bulletin, 2023. Verfasst von Siwei Xue und 8 weiteren Autor:innen. Beteiligte Einrichtung: Shanghai Key Laboratory for Future Computing Hardware and System, School of Microelectronics, Fudan University, Shanghai 200433, China. Hauptthemen laut MEDLINE: Molybdenum, Neural Networks, Computer. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 23 Literaturangaben. Gefördert von: Natural Science Foundation of Shanghai Municipality, National Natural Science Foundation of China, Science and Technology Commission of Shanghai Municipality.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Neuromorphic computing enables efficient processing of data-intensive tasks, but requires numerous artificial synapses and neurons for certain functions, which leads to bulky systems and energy challenges. Achieving functionality with fewer synapses and neurons will facilitate integration density and computility. Two-dimensional (2D) materials exhibit potential for artificial synapses, including diverse biomimetic plasticity and efficient computing. Considering the complexity of neuron circuits and the maturity of complementary metal-oxide-semiconductor (CMOS), hybrid integration is attractive. Here, we demonstrate a hybrid neuromorphic hardware with 2D MoS2 synaptic arrays and CMOS neural circuitry integrated on board. With the joint benefit of hybrid integration, frequency coding and feature extraction, a total cost of twelve MoS2 synapses, three CMOS neurons, combined with digital/analogue converter enables alphabetic and numeric recognition. MoS2 synapses exhibit progressively tunable weight plasticity, CMOS neurons integrate and fire frequency-encoded spikes to display the target characters. The synapse- and neuron-saving hybrid hardware exhibits a competitive accuracy of 98.8% and single recognition energy consumption of 11.4 μW. This work provides a viable solution for building neuromorphic hardware with high compactness and computility.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MolybdenumNeural Networks, ComputerNeuronsSynapsesSemiconductorsOxides

Stichwörter der Autor:innen

2D material synapseCMOS neuronHybrid hardwareNeuromorphic computing

Substanzen

MolybdenumOxides

Autor:innen und Institutionen

  • Shanghai Key Laboratory for Future Computing Hardware and System, School of Microelectronics, Fudan University, Shanghai 200433, China.
  • Shanghai Key Laboratory for Future Computing Hardware and System, School of Microelectronics, Fudan University, Shanghai 200433, China. Electronic address: sy_wang@fudan.edu.cn.
  • State Key Laboratory of ASIC and System, School of Information Science and Technology, Fudan University, Shanghai 200433, China.
  • Shanghai Key Laboratory for Future Computing Hardware and System, School of Microelectronics, Fudan University, Shanghai 200433, China. Electronic address: shunlima@fudan.edu.cn.
  • Shanghai Key Laboratory for Future Computing Hardware and System, School of Microelectronics, Fudan University, Shanghai 200433, China; Frontier Institute of Chip and System & Qizhi Institute, Fudan University, Shanghai…

Förderung

  • Natural Science Foundation of Shanghai Municipality
  • National Natural Science Foundation of China
  • Science and Technology Commission of Shanghai Municipality
  • National Key Research and Development Program of China
  • China Postdoctoral Science Foundation

Zitieren

APA

Xue, S., Wang, S., Wu, T., Di, Z., Xu, N., Sun, Y., Zeng, C., Ma, S., & Zhou, P. (2023). Hybrid neuromorphic hardware with sparing 2D synapse and CMOS neuron for character recognition. Science Bulletin. https://doi.org/10.1016/j.scib.2023.09.006

Vancouver

Xue S, Wang S, Wu T, Di Z, Xu N, Sun Y, et al. Hybrid neuromorphic hardware with sparing 2D synapse and CMOS neuron for character recognition. Science Bulletin. 2023. doi:10.1016/j.scib.2023.09.006

BibTeX
@article{Xue2023,
  author = {Xue, Siwei and Wang, Shuiyuan and Wu, Tianxiang and Di, Ziye and Xu, Nuo and Sun, Yibo and Zeng, Chaofan and Ma, Shunli and Zhou, Peng},
  title = {Hybrid neuromorphic hardware with sparing 2D synapse and CMOS neuron for character recognition},
  journal = {Science Bulletin},
  year = {2023},
  doi = {10.1016/j.scib.2023.09.006}
}
Kostenlos starten →