Skip to main content
Journal cover image

Efficient neural network using pointwise convolution kernels with linear phase constraint

Publication ,  Journal Article
Liang, F; Tian, Z; Dong, M; Cheng, S; Sun, L; Li, H; Chen, Y; Zhang, G
Published in: Neurocomputing
January 29, 2021

In current efficient convolutional neural networks, 1 × 1 convolution is widely used. However, the amount of computation and the number of parameters of 1 × 1 convolution layers account for a large part of these neural network models. In this paper, we propose to use linear-phase pointwise convolution kernels (LPPC kernels) to reduce the computational complexities and storage costs of these neural networks. We design four types of LPPC kernels based on the parity of the number of input channels and symmetry of the weights of the pointwise convolution kernel. Experimental results show that Type-I LPPC kernels can compress some popular networks better with a small reduction in accuracy than the other types of LPPC kernels. The LPPC kernels can be used as new 1 × 1 convolution kernels to design efficient neural network architectures in the future. Moreover, the LPPC kernels are friendly to low-power hardware accelerator design to achieve lower memory access cost and smaller model size.

Duke Scholars

Published In

Neurocomputing

DOI

EISSN

1872-8286

ISSN

0925-2312

Publication Date

January 29, 2021

Volume

423

Start / End Page

572 / 579

Related Subject Headings

  • Artificial Intelligence & Image Processing
  • 52 Psychology
  • 46 Information and computing sciences
  • 40 Engineering
  • 17 Psychology and Cognitive Sciences
  • 09 Engineering
  • 08 Information and Computing Sciences
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Liang, F., Tian, Z., Dong, M., Cheng, S., Sun, L., Li, H., … Zhang, G. (2021). Efficient neural network using pointwise convolution kernels with linear phase constraint. Neurocomputing, 423, 572–579. https://doi.org/10.1016/j.neucom.2020.10.067
Liang, F., Z. Tian, M. Dong, S. Cheng, L. Sun, H. Li, Y. Chen, and G. Zhang. “Efficient neural network using pointwise convolution kernels with linear phase constraint.” Neurocomputing 423 (January 29, 2021): 572–79. https://doi.org/10.1016/j.neucom.2020.10.067.
Liang F, Tian Z, Dong M, Cheng S, Sun L, Li H, et al. Efficient neural network using pointwise convolution kernels with linear phase constraint. Neurocomputing. 2021 Jan 29;423:572–9.
Liang, F., et al. “Efficient neural network using pointwise convolution kernels with linear phase constraint.” Neurocomputing, vol. 423, Jan. 2021, pp. 572–79. Scopus, doi:10.1016/j.neucom.2020.10.067.
Liang F, Tian Z, Dong M, Cheng S, Sun L, Li H, Chen Y, Zhang G. Efficient neural network using pointwise convolution kernels with linear phase constraint. Neurocomputing. 2021 Jan 29;423:572–579.
Journal cover image

Published In

Neurocomputing

DOI

EISSN

1872-8286

ISSN

0925-2312

Publication Date

January 29, 2021

Volume

423

Start / End Page

572 / 579

Related Subject Headings

  • Artificial Intelligence & Image Processing
  • 52 Psychology
  • 46 Information and computing sciences
  • 40 Engineering
  • 17 Psychology and Cognitive Sciences
  • 09 Engineering
  • 08 Information and Computing Sciences