Zobrazeno 1 - 10
of 70
pro vyhledávání: '"Juhyoung Lee"'
Publikováno v:
IEEE Open Journal of the Solid-State Circuits Society, Vol 2, Pp 259-275 (2022)
Many edge/mobile devices are now able to utilize deep neural networks (DNNs) thanks to the development of mobile DNN accelerators. Mobile DNN accelerators overcame the problems of limited computing resources and battery capacity by realizing energy-e
Externí odkaz:
https://doaj.org/article/9c268a238df84521b4d538f537f53830
Publikováno v:
IEEE Journal of Solid-State Circuits. :1-14
Publikováno v:
IEEE Transactions on Circuits and Systems I: Regular Papers. 69:1494-1506
Publikováno v:
IEEE Transactions on Circuits and Systems I: Regular Papers. 69:1507-1518
Autor:
Juhyoung Lee, Sangyeob Kim, Sangjin Kim, Wooyoung Jo, Ji-Hoon Kim, Donghyeon Han, Hoi-Jun Yoo
Publikováno v:
IEEE Journal of Solid-State Circuits. 57:999-1012
Autor:
Sangjin Kim, Zhiyong Li, Soyeon Um, Wooyoung Jo, Sangwoo Ha, Juhyoung Lee, Sangyeob Kim, Donghyeon Han, Hoi-Jun Yoo
Publikováno v:
2023 IEEE International Solid- State Circuits Conference (ISSCC).
Autor:
Donghyeon Han, Dongseok Im, Gwangtae Park, Youngwoo Kim, Seokchan Song, Juhyoung Lee, Hoi-Jun Yoo
Publikováno v:
IEEE Micro. 42:16-25
Publikováno v:
IEEE Micro. 42:99-107
The authors propose a heterogeneous floating-point (FP) computing architecture to maximize energy efficiency by separately optimize exponent processing and mantissa processing. The proposed exponent-computing-in-memory (ECIM) architecture and mantiss
Publikováno v:
2022 IEEE Asian Solid-State Circuits Conference (A-SSCC).
Autor:
Seokchan Song, Juhyoung Lee, Hoi-Jun Yoo, Donghyeon Han, Dongseok Im, Youngwoo Kim, Gwangtae Park
Publikováno v:
IEEE Journal of Solid-State Circuits. 56:2858-2869
This article presents HNPU, which is an energy-efficient deep neural network (DNN) training processor by adopting algorithm-hardware co-design. The HNPU supports stochastic dynamic fixed-point representation and layer-wise adaptive precision searchin