logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

Distribution-modulated binary neural network for image classification

delete2025-07-18
delete0
PRE
AI
Y
Yingcheng Lin
Y
Yuxiao Wang
R
Rui Ding
H
Haijun Liu *
X
Xichuan Zhou
DOI：10.1016/j.imavis.2025.105646delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• Providing a new training strategy for binary neural networks. • Designing a new regularization term to effectively eliminate dead weights. • Providing a well-designed approximation of a peak-shaped derivative. • Achieving SOTA performances by addressing dead weights and susceptible weights.
Keywords:
binary neural networks
regularization term
dead weights
susceptible weights
peak-shaped derivative

Journal

Image and Vision Computing cover
Image and Vision Computing
IF:
4.2
Papers:
4.0K
Citations:
6.7K

Organization

No organization information available