Knowledge Commons of Institute of Automation,CAS
Towards Automatic Model Compression via A Unified Two-Stage Framework | |
Weihan Chen1,2![]() ![]() ![]() | |
发表期刊 | Pattern Recognition (PR)
![]() |
2023 | |
卷号 | 140页码:109527 |
摘要 | Deep Neural Networks have become ubiquitous in various domains. Meanwhile, the problems of massive storage and computation costs have hindered the deploy- ment of these models to real-world applications. This paper proposes a novel and unified two-stage framework for automatic model compression. To determine the compression ratio of each layer, we improve the optimization from two aspects. First, to predict the performance of each compression policy, we propose Dynamic BN, which improves the correlation significantly with little computation overhead. Second, to search for the compression ratio allocation, we propose an efficient and hyperparameter-free solving algorithm based on the proposed Hessian matrix approximation and Knapsack problem reformulation. Moreover, comprehensive experiments and analyses are conducted on the CIFAR-100&ImageNet datasets and various network architectures to demonstrate its performance advantages over existing model compression methods under the quantization-only, pruning-only, and pruning-quantization settings. |
关键词 | Deep Neural Networks Model Compression Quantization Pruning |
收录类别 | SCI |
语种 | 英语 |
七大方向——子方向分类 | AI芯片与智能计算 |
国重实验室规划方向分类 | 智能计算与学习 |
是否有论文关联数据集需要存交 | 否 |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/52076 |
专题 | 复杂系统认知与决策实验室_高效智能计算与学习 |
通讯作者 | Jian Cheng |
作者单位 | 1.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, 100049, China 2.Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China 3.School of Future Technology, University of Chinese Academy of Sciences, Beijing, 100049, China 4.AiRiA |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Weihan Chen,Peisong Wang,Jian Cheng. Towards Automatic Model Compression via A Unified Two-Stage Framework[J]. Pattern Recognition (PR),2023,140:109527. |
APA | Weihan Chen,Peisong Wang,&Jian Cheng.(2023).Towards Automatic Model Compression via A Unified Two-Stage Framework.Pattern Recognition (PR),140,109527. |
MLA | Weihan Chen,et al."Towards Automatic Model Compression via A Unified Two-Stage Framework".Pattern Recognition (PR) 140(2023):109527. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Towards Automatic Mo(765KB) | 期刊论文 | 作者接受稿 | 开放获取 | CC BY-NC-SA | 浏览 |
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