使用参数修剪的数据集蒸馏

IF 0.4 4区计算机科学 Q4 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Ieice Transactions on Fundamentals of Electronics Communications and Computer Sciences Pub Date : 2023-01-01 DOI:10.1587/transfun.2023eal2053

Guang LI, Ren TOGO, Takahiro OGAWA, Miki HASEYAMA

{"title":"使用参数修剪的数据集蒸馏","authors":"Guang LI, Ren TOGO, Takahiro OGAWA, Miki HASEYAMA","doi":"10.1587/transfun.2023eal2053","DOIUrl":null,"url":null,"abstract":"In this study, we propose a novel dataset distillation method based on parameter pruning. The proposed method can synthesize more robust distilled datasets and improve distillation performance by pruning difficult-to-match parameters during the distillation process. Experimental results on two benchmark datasets show the superiority of the proposed method.","PeriodicalId":55003,"journal":{"name":"Ieice Transactions on Fundamentals of Electronics Communications and Computer Sciences","volume":null,"pages":null},"PeriodicalIF":0.4000,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Dataset Distillation Using Parameter Pruning\",\"authors\":\"Guang LI, Ren TOGO, Takahiro OGAWA, Miki HASEYAMA\",\"doi\":\"10.1587/transfun.2023eal2053\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this study, we propose a novel dataset distillation method based on parameter pruning. The proposed method can synthesize more robust distilled datasets and improve distillation performance by pruning difficult-to-match parameters during the distillation process. Experimental results on two benchmark datasets show the superiority of the proposed method.\",\"PeriodicalId\":55003,\"journal\":{\"name\":\"Ieice Transactions on Fundamentals of Electronics Communications and Computer Sciences\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.4000,\"publicationDate\":\"2023-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Ieice Transactions on Fundamentals of Electronics Communications and Computer Sciences\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1587/transfun.2023eal2053\",\"RegionNum\":4,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ieice Transactions on Fundamentals of Electronics Communications and Computer Sciences","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1587/transfun.2023eal2053","RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE","Score":null,"Total":0}

引用次数: 0

摘要

在这项研究中，我们提出了一种新的基于参数修剪的数据集蒸馏方法。该方法通过对蒸馏过程中难以匹配的参数进行修剪，可以合成更鲁棒的蒸馏数据集，提高蒸馏性能。在两个基准数据集上的实验结果表明了该方法的优越性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Dataset Distillation Using Parameter Pruning

In this study, we propose a novel dataset distillation method based on parameter pruning. The proposed method can synthesize more robust distilled datasets and improve distillation performance by pruning difficult-to-match parameters during the distillation process. Experimental results on two benchmark datasets show the superiority of the proposed method.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Ieice Transactions on Fundamentals of Electronics Communications and Computer Sciences 工程技术-工程：电子与电气

CiteScore

1.10

自引率

20.00%

发文量

137

审稿时长

3.9 months

期刊介绍： Includes reports on research, developments, and examinations performed by the Society''s members for the specific fields shown in the category list such as detailed below, the contents of which may advance the development of science and industry: (1) Reports on new theories, experiments with new contents, or extensions of and supplements to conventional theories and experiments. (2) Reports on development of measurement technology and various applied technologies. (3) Reports on the planning, design, manufacture, testing, or operation of facilities, machinery, parts, materials, etc. (4) Presentation of new methods, suggestion of new angles, ideas, systematization, software, or any new facts regarding the above.