函数预定义核:一种减少CNN计算的方法

Yuta Inouchi, Hayato Yamaki, Shinobu Miwa, Tomoaki Tsumura
{"title":"函数预定义核:一种减少CNN计算的方法","authors":"Yuta Inouchi, Hayato Yamaki, Shinobu Miwa, Tomoaki Tsumura","doi":"10.1109/PACRIM47961.2019.8985122","DOIUrl":null,"url":null,"abstract":"Convolutional Neural Networks (CNNs) have achieved high classification accuracy in image recognition, and now, they are widely used for numerous applications. For higher accuracy or more advanced applications, CNNs need to consume tremendous computational resources and time. Hence, many studies for reducing the computational cost of CNNs are actively being conducted. However, many previous methods for reducing the computational cost lead to a non-negligible loss in output accuracy. Therefore, it is still a challenge to reduce the computational cost of CNNs with keeping the output accuracy high. In this paper, we propose a novel concept \"Functionally-Predefined Kernel\" to reduce the computational cost for CNN training and discuss the potential of computation reuse to reduce the computational cost for CNN inference. Our experimental results show that the number of parameters to be trained can be significantly reduced by utilizing Functionally-Predefined Kernels without accuracy loss. In addition, we revealed that CNN’s inference process includes many convolution operations with the same inputs and computation reuse, therefore, has high affinity to CNN computation.","PeriodicalId":152556,"journal":{"name":"2019 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing (PACRIM)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Functionally-Predefined Kernel: a Way to Reduce CNN Computation\",\"authors\":\"Yuta Inouchi, Hayato Yamaki, Shinobu Miwa, Tomoaki Tsumura\",\"doi\":\"10.1109/PACRIM47961.2019.8985122\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Convolutional Neural Networks (CNNs) have achieved high classification accuracy in image recognition, and now, they are widely used for numerous applications. For higher accuracy or more advanced applications, CNNs need to consume tremendous computational resources and time. Hence, many studies for reducing the computational cost of CNNs are actively being conducted. However, many previous methods for reducing the computational cost lead to a non-negligible loss in output accuracy. Therefore, it is still a challenge to reduce the computational cost of CNNs with keeping the output accuracy high. In this paper, we propose a novel concept \\\"Functionally-Predefined Kernel\\\" to reduce the computational cost for CNN training and discuss the potential of computation reuse to reduce the computational cost for CNN inference. Our experimental results show that the number of parameters to be trained can be significantly reduced by utilizing Functionally-Predefined Kernels without accuracy loss. In addition, we revealed that CNN’s inference process includes many convolution operations with the same inputs and computation reuse, therefore, has high affinity to CNN computation.\",\"PeriodicalId\":152556,\"journal\":{\"name\":\"2019 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing (PACRIM)\",\"volume\":\"14 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2019-08-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2019 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing (PACRIM)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/PACRIM47961.2019.8985122\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing (PACRIM)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PACRIM47961.2019.8985122","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

摘要

卷积神经网络(Convolutional Neural Networks, cnn)在图像识别中取得了很高的分类精度,目前被广泛应用于众多领域。为了获得更高的精度或更高级的应用,cnn需要消耗大量的计算资源和时间。因此,许多降低cnn计算成本的研究正在积极进行。然而,许多先前减少计算成本的方法导致输出精度的不可忽略的损失。因此,如何在保持高输出精度的同时降低cnn的计算成本仍然是一个挑战。在本文中,我们提出了一个新的概念“功能预定义内核”来降低CNN训练的计算成本,并讨论了计算重用的潜力,以降低CNN推理的计算成本。我们的实验结果表明,使用功能预定义核可以在不损失精度的情况下显著减少待训练参数的数量。此外,我们发现CNN的推理过程包含许多相同输入和计算重用的卷积操作,因此对CNN计算具有很高的亲和力。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Functionally-Predefined Kernel: a Way to Reduce CNN Computation
Convolutional Neural Networks (CNNs) have achieved high classification accuracy in image recognition, and now, they are widely used for numerous applications. For higher accuracy or more advanced applications, CNNs need to consume tremendous computational resources and time. Hence, many studies for reducing the computational cost of CNNs are actively being conducted. However, many previous methods for reducing the computational cost lead to a non-negligible loss in output accuracy. Therefore, it is still a challenge to reduce the computational cost of CNNs with keeping the output accuracy high. In this paper, we propose a novel concept "Functionally-Predefined Kernel" to reduce the computational cost for CNN training and discuss the potential of computation reuse to reduce the computational cost for CNN inference. Our experimental results show that the number of parameters to be trained can be significantly reduced by utilizing Functionally-Predefined Kernels without accuracy loss. In addition, we revealed that CNN’s inference process includes many convolution operations with the same inputs and computation reuse, therefore, has high affinity to CNN computation.
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
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信