• Complex
  • Title
  • Keyword
  • Abstract
  • Scholars
  • Journal
  • ISSN
  • Conference
搜索

Author:

Xu, Jianrong (Xu, Jianrong.) | Diao, Boyu (Diao, Boyu.) | Cui, Bifeng (Cui, Bifeng.) | Yang, Kang (Yang, Kang.) | Li, Chao (Li, Chao.) | Hong, Hailong (Hong, Hailong.)

Indexed by:

CPCI-S EI Scopus

Abstract:

Deep learning has achieved impressive results in many areas, but the deployment of edge intelligent devices is still very slow. To solve this problem, we propose a novel compression and acceleration method based on data distribution characteristics for deep neural networks, namely Pruning Filter via Gaussian Distribution Feature (PFGDF). Compared with previous advanced pruning methods, PFGDF compresses the model by filters with insignificance in distribution, regardless of the contribution and sensitivity information of the convolution filter. PFGDF is significantly different from weight sparsification pruning because it does not require the special accelerated library to process the sparse weight matrix and introduces no more extra parameters. The pruning process of PFGDF is automated. Furthermore, the model compressed by PFGDF can restore the same performance as the uncompressed model. We evaluate PFGDF through extensive experiments, on CIFAR-10, PFGDF compresses the convolution filter on VGG-16 by 66:62% with more than 90% parameter reduced, while the inference time is accelerated by 83:73% on Huawei MATE 10.

Keyword:

neural network compression network pruning deep learning

Author Community:

  • [ 1 ] [Xu, Jianrong]Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China
  • [ 2 ] [Diao, Boyu]Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China
  • [ 3 ] [Yang, Kang]Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China
  • [ 4 ] [Li, Chao]Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China
  • [ 5 ] [Hong, Hailong]Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China
  • [ 6 ] [Xu, Jianrong]Beijing Univ Technol, Fac Informat Technol, Beijing, Peoples R China
  • [ 7 ] [Cui, Bifeng]Beijing Univ Technol, Fac Informat Technol, Beijing, Peoples R China

Reprint Author's Address:

Show more details

Related Keywords:

Related Article:

Source :

2022 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN)

ISSN: 2161-4393

Year: 2022

Cited Count:

WoS CC Cited Count: 3

SCOPUS Cited Count: 4

ESI Highly Cited Papers on the List: 0 Unfold All

WanFang Cited Count:

Chinese Cited Count:

30 Days PV: 9

Affiliated Colleges:

Online/Total:618/10811262
Address:BJUT Library(100 Pingleyuan,Chaoyang District,Beijing 100124, China Post Code:100124) Contact Us:010-67392185
Copyright:BJUT Library Technical Support:Beijing Aegean Software Co., Ltd.