English

NAM: Normalization-based Attention Module

Computer Vision and Pattern Recognition 2021-11-25 v1 Image and Video Processing

Abstract

Recognizing less salient features is the key for model compression. However, it has not been investigated in the revolutionary attention mechanisms. In this work, we propose a novel normalization-based attention module (NAM), which suppresses less salient weights. It applies a weight sparsity penalty to the attention modules, thus, making them more computational efficient while retaining similar performance. A comparison with three other attention mechanisms on both Resnet and Mobilenet indicates that our method results in higher accuracy. Code for this paper can be publicly accessed at https://github.com/Christian-lyc/NAM.

Keywords

Cite

@article{arxiv.2111.12419,
  title  = {NAM: Normalization-based Attention Module},
  author = {Yichao Liu and Zongru Shao and Yueyang Teng and Nico Hoffmann},
  journal= {arXiv preprint arXiv:2111.12419},
  year   = {2021}
}

Comments

3 pages, 2 figures, 2 tables, 2 tables in the appendix

R2 v1 2026-06-24T07:50:20.834Z