Multiple Algorithms Against Multiple Hardware Architectures: Data-Driven Exploration on Deep Convolution Neural Network - Network and Parallel Computing Access content directly
Conference Papers Year : 2019

Multiple Algorithms Against Multiple Hardware Architectures: Data-Driven Exploration on Deep Convolution Neural Network

Chongyang Xu
  • Function : Author
  • PersonId : 1161636
Zhongzhi Luan
  • Function : Author
  • PersonId : 1023705
Lan Gao
  • Function : Author
  • PersonId : 1161637
Rui Wang
  • Function : Author
  • PersonId : 1161638
Han Zhang
  • Function : Author
  • PersonId : 1161639
Yi Liu
  • Function : Author
  • PersonId : 1161641
Depei Qian
  • Function : Author
  • PersonId : 1023707

Abstract

With the rapid development of deep learning (DL), various convolution neural network (CNN) models have been developed. Moreover, to execute different DL workloads efficiently, many accelerators have been proposed. To guide the design of both CNN models and hardware architectures for a high-performance inference system, we choose five types of CNN models and test them on six processors and measure three metrics. With our experiments, we get two observations and conduct two insights for the design of CNN algorithms and hardware architectures.
Fichier principal
Vignette du fichier
486810_1_En_36_Chapter.pdf (1.09 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03770535 , version 1 (06-09-2022)

Licence

Attribution

Identifiers

Cite

Chongyang Xu, Zhongzhi Luan, Lan Gao, Rui Wang, Han Zhang, et al.. Multiple Algorithms Against Multiple Hardware Architectures: Data-Driven Exploration on Deep Convolution Neural Network. 16th IFIP International Conference on Network and Parallel Computing (NPC), Aug 2019, Hohhot, China. pp.371-375, ⟨10.1007/978-3-030-30709-7_36⟩. ⟨hal-03770535⟩
9 View
20 Download

Altmetric

Share

Gmail Facebook X LinkedIn More