零标签和多标签图像的准二分类器

Shuai Liao, E. Gavves, Changyong Oh, Cees G. M. Snoek
{"title":"零标签和多标签图像的准二分类器","authors":"Shuai Liao, E. Gavves, Changyong Oh, Cees G. M. Snoek","doi":"10.1109/ICPR48806.2021.9412933","DOIUrl":null,"url":null,"abstract":"The softmax and binary classifier are commonly preferred for image classification applications. However, as softmax is specifically designed for categorical classification, it assumes each image has just one class label. This limits its applicability for problems where the number of labels does not equal one, most notably zero- and multi-label problems. In these challenging settings, binary classifiers are, in theory, better suited. However, as they ignore the correlation between classes, they are not as accurate and scalable in practice. In this paper, we start from the observation that the only difference between binary and softmax classifiers is their normalization function. Specifically, while the binary classifier self-normalizes its score, the softmax classifier combines the scores from all classes before normalisation. On the basis of this observation we introduce a normalization function that is learnable, constant, and shared between classes and data points. By doing so, we arrive at a new type of binary classifier that we coin quasibinary classifier. We show in a variety of image classification settings, and on several datasets, that quasibinary classifiers are considerably better in classification settings where regular binary and softmax classifiers suffer, including zero-label and multi-label classification. What is more, we show that quasibinary classifiers yield well-calibrated probabilities allowing for direct and reliable comparisons, not only between classes but also between data points.","PeriodicalId":6783,"journal":{"name":"2020 25th International Conference on Pattern Recognition (ICPR)","volume":"100 1","pages":"8743-8750"},"PeriodicalIF":0.0000,"publicationDate":"2021-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Quasibinary Classifier for Images with Zero and Multiple Labels\",\"authors\":\"Shuai Liao, E. Gavves, Changyong Oh, Cees G. M. Snoek\",\"doi\":\"10.1109/ICPR48806.2021.9412933\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The softmax and binary classifier are commonly preferred for image classification applications. However, as softmax is specifically designed for categorical classification, it assumes each image has just one class label. This limits its applicability for problems where the number of labels does not equal one, most notably zero- and multi-label problems. In these challenging settings, binary classifiers are, in theory, better suited. However, as they ignore the correlation between classes, they are not as accurate and scalable in practice. In this paper, we start from the observation that the only difference between binary and softmax classifiers is their normalization function. Specifically, while the binary classifier self-normalizes its score, the softmax classifier combines the scores from all classes before normalisation. On the basis of this observation we introduce a normalization function that is learnable, constant, and shared between classes and data points. By doing so, we arrive at a new type of binary classifier that we coin quasibinary classifier. We show in a variety of image classification settings, and on several datasets, that quasibinary classifiers are considerably better in classification settings where regular binary and softmax classifiers suffer, including zero-label and multi-label classification. What is more, we show that quasibinary classifiers yield well-calibrated probabilities allowing for direct and reliable comparisons, not only between classes but also between data points.\",\"PeriodicalId\":6783,\"journal\":{\"name\":\"2020 25th International Conference on Pattern Recognition (ICPR)\",\"volume\":\"100 1\",\"pages\":\"8743-8750\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-01-10\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2020 25th International Conference on Pattern Recognition (ICPR)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICPR48806.2021.9412933\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 25th International Conference on Pattern Recognition (ICPR)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICPR48806.2021.9412933","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

softmax和二进制分类器通常是图像分类应用的首选。然而,由于softmax是专门为分类分类设计的,它假设每个图像只有一个类标签。这限制了它对标签数量不等于1的问题的适用性,尤其是零标签和多标签问题。在这些具有挑战性的环境中,理论上,二元分类器更适合。然而,由于它们忽略了类之间的相关性,因此在实践中它们不那么准确和可扩展。在本文中,我们从观察到二进制分类器和softmax分类器之间的唯一区别是它们的归一化函数开始。具体来说,当二元分类器自归一化其分数时,softmax分类器在归一化之前将所有类别的分数组合在一起。在此观察的基础上,我们引入了一个可学习的、恒定的、在类和数据点之间共享的归一化函数。通过这样做,我们得到了一种新的二元分类器,即拟二元分类器。我们在各种图像分类设置和几个数据集上显示,准二元分类器在常规二元和softmax分类器遭受损失的分类设置中表现得更好,包括零标签和多标签分类。更重要的是,我们表明准二元分类器产生良好校准的概率,允许直接和可靠的比较,不仅在类之间,而且在数据点之间。
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
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Quasibinary Classifier for Images with Zero and Multiple Labels
The softmax and binary classifier are commonly preferred for image classification applications. However, as softmax is specifically designed for categorical classification, it assumes each image has just one class label. This limits its applicability for problems where the number of labels does not equal one, most notably zero- and multi-label problems. In these challenging settings, binary classifiers are, in theory, better suited. However, as they ignore the correlation between classes, they are not as accurate and scalable in practice. In this paper, we start from the observation that the only difference between binary and softmax classifiers is their normalization function. Specifically, while the binary classifier self-normalizes its score, the softmax classifier combines the scores from all classes before normalisation. On the basis of this observation we introduce a normalization function that is learnable, constant, and shared between classes and data points. By doing so, we arrive at a new type of binary classifier that we coin quasibinary classifier. We show in a variety of image classification settings, and on several datasets, that quasibinary classifiers are considerably better in classification settings where regular binary and softmax classifiers suffer, including zero-label and multi-label classification. What is more, we show that quasibinary classifiers yield well-calibrated probabilities allowing for direct and reliable comparisons, not only between classes but also between data points.
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
期刊最新文献
Trajectory representation learning for Multi-Task NMRDP planning Semantic Segmentation Refinement Using Entropy and Boundary-guided Monte Carlo Sampling and Directed Regional Search A Randomized Algorithm for Sparse Recovery An Empirical Bayes Approach to Topic Modeling To Honor our Heroes: Analysis of the Obituaries of Australians Killed in Action in WWI and WWII
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1