Progressive Channel-Shrinking Network

doi:10.1109/TMM.2023.3291197

Progressive Channel-Shrinking Network

2024-01-01

0

OA

AI

DOI：10.1109/TMM.2023.3291197

Original

Original request for help

Save

Abstract

En 中文

Currently, salience-based channel pruning makes continuous breakthroughs in network compression. In the realization, the salience mechanism is used as a metric of channel salience to guide pruning. Therefore, salience-based channel pruning can dynamically adjust the channel width at run-time, which provides a flexible pruning scheme. However, there are two problems emerging: a gating function is often needed to truncate the specific salience entries to zero, which destabilizes the forward propagation; dynamic architecture brings more cost for indexing in inference which bottlenecks the inference speed. In this article, we propose a Progressive Channel-Shrinking (PCS) method to compress the selected salience entries at run-time instead of roughly approximating them to zero. We also propose a Running Shrinking Policy to provide a testing-static pruning scheme that can reduce the memory access cost for filter indexing. We evaluate our method on ImageNet and CIFAR10 datasets over two prevalent networks: ResNet and VGG, and demonstrate that our PCS outperforms all baselines and achieves state-of-the-art in terms of compression-performance tradeoff. Moreover, we observe a significant and practical acceleration of inference. The code is available at https://github.com/JianhongPan-VLG/Progressive.Channel-Shrinking.Network.

Keywords:

Training

Indexing

Convolution

Costs

Generators

Feature extraction

Testing

Progressive

network shrinking

Progressive Channel-Shrinking Network

Abstract

Journal

Organization