中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
PHD-NAS: Preserving helpful data to promote Neural Architecture Search

文献类型：期刊论文

作者Lu, Shun1,2; Hu, Yu1,2; Yang, Longxing1,2; Mei, Jilin1; Sun, Zihao1,2; Tan, Jianchao3; Song, Chengru3
刊名NEUROCOMPUTING
出版日期2024-06-28
卷号587页码:14
关键词Neural architecture search Dataset optimization Forgetting events and remembering events
ISSN号0925-2312
DOI10.1016/j.neucom.2024.127646
英文摘要Neural Architecture Search (NAS) has achieved promising results in many domains. However, the enormous computational burden consumed by the NAS procedure significantly hinders its application. Existing works focus on mitigating the search cost by either designing a more efficient algorithm or searching in an elaborately designed search space, heavily relying on expert experience and domain knowledge. We notice that few works focus on dataset optimization for NAS, however, the truth is that not all samples are essential for the search process, which can be omitted actually. Therefore, we propose to only preserve helpful data for the supernet training to improve the efficiency. Specifically, we compute the forgetting and remembering events for each sample during the supernet training to determine the data importance. Samples that the supernet has predicted correctly in consecutive epochs have low importance and will be gradually removed from the dataset during training. We further formulate our method into a unified cycled -learning framework for jointly optimizing proxy dataset and architecture search. By combining with different algorithms, we demonstrate that our framework can find architectures with comparable performance using much less training data and search time in various search spaces and benchmarks, validating the effectiveness of our method.
资助项目National Key R&D Program of China[2018AAA0102701] ; National Natural Science Foundation of China[62176250]
WOS研究方向Computer Science
语种英语
WOS记录号WOS:001231918800001
出版者ELSEVIER
源URL[http://119.78.100.204/handle/2XEOYT63/40084]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Hu, Yu
作者单位1.Chinese Acad Sci, Inst Comp Technol, Res Ctr Intelligent Comp Syst, Beijing 100190, Peoples R China
2.Univ Chinese Acad Sci, Sch Comp Sci & Technol, Beijing 100049, Peoples R China
3.Kuaishou Technol, Beijing 100085, Peoples R China
推荐引用方式
GB/T 7714		 Lu, Shun,Hu, Yu,Yang, Longxing,et al. PHD-NAS: Preserving helpful data to promote Neural Architecture Search[J]. NEUROCOMPUTING,2024,587:14.
APA Lu, Shun.,Hu, Yu.,Yang, Longxing.,Mei, Jilin.,Sun, Zihao.,...&Song, Chengru.(2024).PHD-NAS: Preserving helpful data to promote Neural Architecture Search.NEUROCOMPUTING,587,14.
MLA Lu, Shun,et al."PHD-NAS: Preserving helpful data to promote Neural Architecture Search".NEUROCOMPUTING 587(2024):14.

入库方式： OAI收割

来源：计算技术研究所

浏览0
下载0
收藏0
其他版本

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。