SOTAVerified

NAM: Normalization-based Attention Module

2021-11-24NeurIPS Workshop ImageNet_PPF 2021Code Available1· sign in to hype

Yichao Liu, Zongru Shao, Yueyang Teng, Nico Hoffmann

Code Available — Be the first to reproduce this paper.

Reproduce

Code

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.

Tasks

Reproductions