Pruning-aware Sparse Regularization for Network Pruning

TL;DR

This paper introduces MaskSparsity, a pruning-aware sparse regularization method that selectively applies regularization to pruned filters, improving neural network pruning efficiency without accuracy loss.

Contribution

Proposes MaskSparsity, a novel pruning method that applies regularization only to selected filters, enhancing pruning effectiveness and reducing performance degradation.

Findings

Achieves 63.03% FLOPs reduction on ResNet-110 with no accuracy loss.

Reduces over 51.07% FLOPs on ResNet-50 with minimal accuracy drop.

Code integrated into EasyPruner toolkit for practical use.

Abstract

Structural neural network pruning aims to remove the redundant channels in the deep convolutional neural networks (CNNs) by pruning the filters of less importance to the final output accuracy. To reduce the degradation of performance after pruning, many methods utilize the loss with sparse regularization to produce structured sparsity. In this paper, we analyze these sparsity-training-based methods and find that the regularization of unpruned channels is unnecessary. Moreover, it restricts the network's capacity, which leads to under-fitting. To solve this problem, we propose a novel pruning method, named MaskSparsity, with pruning-aware sparse regularization. MaskSparsity imposes the fine-grained sparse regularization on the specific filters selected by a pruning mask, rather than all the filters of the model. Before the fine-grained sparse regularization of MaskSparity, we can use many methods to get the pruning mask, such as running the global sparse regularization. MaskSparsity achieves 63.03%-FLOPs reduction on ResNet-110 by removing 60.34% of the parameters, with no top-1 accuracy loss on CIFAR-10. On ILSVRC-2012, MaskSparsity reduces more than 51.07% FLOPs on ResNet-50, with only a loss of 0.76% in the top-1 accuracy. The code is released at https://github.com/CASIA-IVA-Lab/MaskSparsity. Moreover, we have integrated the code of MaskSparity into a PyTorch pruning toolkit, EasyPruner, at https://gitee.com/casia_iva_engineer/easypruner.