EUVIMEDEUVIMED entdecken
Crossref

An FPGA-based bit-level weight sparsity and mixed-bit accelerator for neural networks

Xianghong Hu, Shansen Fu, Yuanmiao Lin, Xueming Li, Chaoming Yang, Rongfeng Li, Hongmin Huang, Shuting Cai, Xiaoming Xiong

Journal of Systems Architecture · 2025-09 · 166 · 103463

DOI: 10.1016/j.sysarc.2025.103463 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Systems Architecture, 2025-09. Verfasst von Xianghong Hu und 8 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 42 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Basic and Applied Basic Research Foundation of Guangdong Province, Science and Technology Planning Project of Guangzhou.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Kein Abstract verfügbar.

Förderung

  • National Natural Science Foundation of China
  • Basic and Applied Basic Research Foundation of Guangdong Province
  • Science and Technology Planning Project of Guangzhou
  • Guangzhou Municipal Science and Technology Project

Zitieren

APA

Hu, X., Fu, S., Lin, Y., Li, X., Yang, C., Li, R., Huang, H., Cai, S., & Xiong, X. (2025). An FPGA-based bit-level weight sparsity and mixed-bit accelerator for neural networks. Journal of Systems Architecture, 166, 103463. https://doi.org/10.1016/j.sysarc.2025.103463

Vancouver

Hu X, Fu S, Lin Y, Li X, Yang C, Li R, et al. An FPGA-based bit-level weight sparsity and mixed-bit accelerator for neural networks. Journal of Systems Architecture. 2025;166:103463. doi:10.1016/j.sysarc.2025.103463

BibTeX
@article{Hu2025,
  author = {Hu, Xianghong and Fu, Shansen and Lin, Yuanmiao and Li, Xueming and Yang, Chaoming and Li, Rongfeng and Huang, Hongmin and Cai, Shuting and Xiong, Xiaoming},
  title = {An FPGA-based bit-level weight sparsity and mixed-bit accelerator for neural networks},
  journal = {Journal of Systems Architecture},
  year = {2025},
  volume = {166},
  pages = {103463},
  doi = {10.1016/j.sysarc.2025.103463}
}
Kostenlos starten →