Channel Pruning Guided by Classification Loss and Feature Importance

TL;DR

This paper introduces CPLI, a layer-by-layer channel pruning method that considers classification loss and feature importance, leading to more effective model compression.

Contribution

It proposes a novel channel pruning approach that incorporates classification loss and feature importance, improving pruning efficiency and effectiveness.

Findings

Effective on CIFAR-10, ImageNet, and UCF-101 datasets

Outperforms existing layer-by-layer pruning methods

Reduces model complexity while maintaining accuracy

Abstract

In this work, we propose a new layer-by-layer channel pruning method called Channel Pruning guided by classification Loss and feature Importance (CPLI). In contrast to the existing layer-by-layer channel pruning approaches that only consider how to reconstruct the features from the next layer, our approach additionally take the classification loss into account in the channel pruning process. We also observe that some reconstructed features will be removed at the next pruning stage. So it is unnecessary to reconstruct these features. To this end, we propose a new strategy to suppress the influence of unimportant features (i.e., the features will be removed at the next pruning stage). Our comprehensive experiments on three benchmark datasets, i.e., CIFAR-10, ImageNet, and UCF-101, demonstrate the effectiveness of our CPLI method.