Pruning-aware Sparse Regularization for Network Pruning | |
Nan-Fei Jiang1,2 | |
刊名 | Machine Intelligence Research |
2023 | |
卷号 | 20期号:1页码:109-120 |
关键词 | Deep learning convolutional neural network (CNN) model compression and acceleration network pruning regularization |
ISSN号 | 2731-538X |
DOI | 10.1007/s11633-022-1353-0 |
英文摘要 | 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-train ing-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 a 63.03% float point operations (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 of this paper is released at https://github.com/CASIA-IVA-Lab/MaskSparsity. We have also integrated the code into a self-developed PyTorch pruning toolkit, named EasyPruner, at https://gitee.com/casia_iva_engineer/easypruner. |
内容类型 | 期刊论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/50903] |
专题 | 自动化研究所_学术期刊_International Journal of Automation and Computing |
作者单位 | 1.Institute of Automation, Chinese Academy of Sciences, Beijing 100049, China 2.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, China |
推荐引用方式 GB/T 7714 | Nan-Fei Jiang. Pruning-aware Sparse Regularization for Network Pruning[J]. Machine Intelligence Research,2023,20(1):109-120. |
APA | Nan-Fei Jiang.(2023).Pruning-aware Sparse Regularization for Network Pruning.Machine Intelligence Research,20(1),109-120. |
MLA | Nan-Fei Jiang."Pruning-aware Sparse Regularization for Network Pruning".Machine Intelligence Research 20.1(2023):109-120. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论