Layer Pruning for Accelerating Very Deep Neural Networks

TL;DR

This paper introduces an adaptive layer and channel pruning technique for very deep neural networks that reduces parameters by half without sacrificing accuracy, and sometimes even improving it.

Contribution

It presents a novel adaptive pruning method that learns to cut channels and layers dynamically, enhancing efficiency while maintaining or improving performance.

Findings

Reduces model parameters by 50%

Maintains or improves baseline accuracy

Adaptive pruning learns optimal cuts dynamically

Abstract

In this paper, we propose an adaptive pruning method. This method can cut off the channel and layer adaptively. The proportion of the layer and the channel to be cut is learned adaptively. The pruning method proposed in this paper can reduce half of the parameters, and the accuracy will not decrease or even be higher than baseline.