Correlation between Alignment-Uniformity and Performance of Dense Contrastive Representations

J. Moon, Wonjae Kim, E. Choi
{"title":"Correlation between Alignment-Uniformity and Performance of Dense Contrastive Representations","authors":"J. Moon, Wonjae Kim, E. Choi","doi":"10.48550/arXiv.2210.08819","DOIUrl":null,"url":null,"abstract":"Recently, dense contrastive learning has shown superior performance on dense prediction tasks compared to instance-level contrastive learning. Despite its supremacy, the properties of dense contrastive representations have not yet been carefully studied. Therefore, we analyze the theoretical ideas of dense contrastive learning using a standard CNN and straightforward feature matching scheme rather than propose a new complex method. Inspired by the analysis of the properties of instance-level contrastive representations through the lens of alignment and uniformity on the hypersphere, we employ and extend the same lens for the dense contrastive representations to analyze their underexplored properties. We discover the core principle in constructing a positive pair of dense features and empirically proved its validity. Also, we introduces a new scalar metric that summarizes the correlation between alignment-and-uniformity and downstream performance. Using this metric, we study various facets of densely learned contrastive representations such as how the correlation changes over single- and multi-object datasets or linear evaluation and dense prediction tasks. The source code is publicly available at: https://github.com/SuperSupermoon/DenseCL-analysis","PeriodicalId":72437,"journal":{"name":"BMVC : proceedings of the British Machine Vision Conference. British Machine Vision Conference","volume":"1 1","pages":"844"},"PeriodicalIF":0.0000,"publicationDate":"2022-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"BMVC : proceedings of the British Machine Vision Conference. British Machine Vision Conference","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.48550/arXiv.2210.08819","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2

Abstract

Recently, dense contrastive learning has shown superior performance on dense prediction tasks compared to instance-level contrastive learning. Despite its supremacy, the properties of dense contrastive representations have not yet been carefully studied. Therefore, we analyze the theoretical ideas of dense contrastive learning using a standard CNN and straightforward feature matching scheme rather than propose a new complex method. Inspired by the analysis of the properties of instance-level contrastive representations through the lens of alignment and uniformity on the hypersphere, we employ and extend the same lens for the dense contrastive representations to analyze their underexplored properties. We discover the core principle in constructing a positive pair of dense features and empirically proved its validity. Also, we introduces a new scalar metric that summarizes the correlation between alignment-and-uniformity and downstream performance. Using this metric, we study various facets of densely learned contrastive representations such as how the correlation changes over single- and multi-object datasets or linear evaluation and dense prediction tasks. The source code is publicly available at: https://github.com/SuperSupermoon/DenseCL-analysis
密集对比表示的对齐均匀性与性能之间的关系
近年来,密集对比学习在密集预测任务上表现出比实例级对比学习更优越的性能。尽管密集对比表征具有至高无上的地位,但其性质尚未被仔细研究。因此,我们使用标准的CNN和简单的特征匹配方案来分析密集对比学习的理论思想,而不是提出一种新的复杂方法。受通过超球上的对齐和均匀性透镜分析实例级对比表示属性的启发,我们对密集对比表示使用并扩展了相同的透镜来分析它们未被探索的属性。我们发现了构造正对密集特征的核心原理,并通过经验证明了其有效性。此外,我们还引入了一个新的标量度量,总结了对准和均匀性与下游性能之间的相关性。使用该度量,我们研究了密集学习对比表示的各个方面,例如单目标和多目标数据集或线性评估和密集预测任务之间的相关性如何变化。源代码可以在:https://github.com/SuperSupermoon/DenseCL-analysis上公开获得
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
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:604180095
Book学术官方微信