EUVIMEDEUVIMED entdecken
Crossref

Programming memristor arrays with arbitrarily high precision for analog computing

Wenhao Song iD, Mingyi Rao iD, Yunning Li, Can Li iD, Ye Zhuo iD, Fuxi Cai iD, Mingche Wu, Wenbo Yin, Zongze Li, Qiang Wei iD, Sangsoo Lee iD, Hengfang Zhu und 9 weitere

Science · 23.02.2024 · 383 (6685) · 903-910

DOI: 10.1126/science.adi9405 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science, erschienen am 23.02.2024. Verfasst von Wenhao Song und 20 weiteren Autor:innen. Beteiligte Einrichtung: Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, USA. Die Arbeit wurde bisher 220-mal zitiert (Crossref) und nennt 51 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In-memory computing represents an effective method for modeling complex physical systems that are typically challenging for conventional computing architectures but has been hindered by issues such as reading noise and writing variability that restrict scalability, accuracy, and precision in high-performance computations. We propose and demonstrate a circuit architecture and programming protocol that converts the analog computing result to digital at the last step and enables low-precision analog devices to perform high-precision computing. We use a weighted sum of multiple devices to represent one number, in which subsequently programmed devices are used to compensate for preceding programming errors. With a memristor system-on-chip, we experimentally demonstrate high-precision solutions for multiple scientific computing tasks while maintaining a substantial power efficiency advantage over conventional digital approaches.

Autor:innen und Institutionen

  • Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, USA.
  • TetraMem Inc., Fremont, CA, USA.
  • Department of Electrical and Computer Engineering, University of Massachusetts, Amherst, MA, USA.
  • Air Force Research Lab, Information Directorate, Rome, NY, USA.

ORCID-Profile: Wenhao Song, Mingyi Rao, Can Li, Ye Zhuo, Fuxi Cai, Qiang Wei, Sangsoo Lee, Mark Barnell, Qing Wu, Peter A Beerel, Mike Shuo-Wei Chen, Ning Ge, Qiangfei Xia, J Joshua Yang

Zitieren

APA

Song, W., Rao, M., Li, Y., Li, C., Zhuo, Y., Cai, F., Wu, M., Yin, W., Li, Z., Wei, Q., Lee, S., Zhu, H., Gong, L., Barnell, M., Wu, Q., Beerel, P. A., Chen, M. S., Ge, N., Hu, M., …, & Yang, J. J. (2024). Programming memristor arrays with arbitrarily high precision for analog computing. Science, 383(6685), 903-910. https://doi.org/10.1126/science.adi9405

Vancouver

Song W, Rao M, Li Y, Li C, Zhuo Y, Cai F, et al. Programming memristor arrays with arbitrarily high precision for analog computing. Science. 2024;383(6685):903-910. doi:10.1126/science.adi9405

BibTeX
@article{Song2024,
  author = {Song, Wenhao and Rao, Mingyi and Li, Yunning and Li, Can and Zhuo, Ye and Cai, Fuxi and Wu, Mingche and Yin, Wenbo and Li, Zongze and Wei, Qiang and Lee, Sangsoo and Zhu, Hengfang and Gong, Lei and Barnell, Mark and Wu, Qing and Beerel, Peter A and Chen, Mike Shuo-Wei and Ge, Ning and Hu, Miao and Xia, Qiangfei and Yang, J Joshua},
  title = {Programming memristor arrays with arbitrarily high precision for analog computing},
  journal = {Science},
  year = {2024},
  volume = {383},
  number = {6685},
  pages = {903--910},
  doi = {10.1126/science.adi9405}
}
Kostenlos starten →