基于动量对比学习的星系形态分类模型

IF 3.3 3区 物理与天体物理 Q2 ASTRONOMY & ASTROPHYSICS Publications of the Astronomical Society of the Pacific Pub Date : 2023-10-01 DOI:10.1088/1538-3873/acf8f7
Guoqiang Shen, Zhiqiang Zou, A-Li Luo, Shuxin Hong, Xiao Kong
{"title":"基于动量对比学习的星系形态分类模型","authors":"Guoqiang Shen, Zhiqiang Zou, A-Li Luo, Shuxin Hong, Xiao Kong","doi":"10.1088/1538-3873/acf8f7","DOIUrl":null,"url":null,"abstract":"Abstract The taxonomy of galaxy morphology plays an important role in astrophysics and provides great help for the study of galaxy evolution. To integrate the advantages of unsupervised learning without labels and supervised learning with high classification accuracy, this paper proposes a galaxy morphology classification model based on a momentum contrastive learning algorithm named Momentum Contrastive Learning Galaxy (MCL-Galaxy), which mainly includes two parts (i) pre-training of the model, where the ResNet_50 backbone network acts as an encoder to learn the galaxy morphology image features, which are stored in the queue and their consistency is ensured by using the momentum contrastive learning algorithm; and (ii) transfer learning, where Mahalanobis distance can assist in improving classification accuracy in downstream tasks where both encoder and queue are transferred. To evaluate the performance of MCL-Galaxy, we use the data set of the Galaxy Zoo challenge project on Kaggle for comparative testing. The experimental results show that the classification accuracy of MCL-Galaxy can reach 90.12%, which is 8.12% higher than the unsupervised state-of-the-art results. Although it is 3.1% lower than the advanced supervised method, it has the advantage of no label and can achieve a higher accuracy rate at the first epoch of classification iteration. This suggests that the gap between unsupervised and supervised representation learning in the field of Galaxy Morphologies classification tasks is well bridged.","PeriodicalId":20820,"journal":{"name":"Publications of the Astronomical Society of the Pacific","volume":"50 1","pages":"0"},"PeriodicalIF":3.3000,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A Galaxy Morphology Classification Model Based on Momentum Contrastive Learning\",\"authors\":\"Guoqiang Shen, Zhiqiang Zou, A-Li Luo, Shuxin Hong, Xiao Kong\",\"doi\":\"10.1088/1538-3873/acf8f7\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract The taxonomy of galaxy morphology plays an important role in astrophysics and provides great help for the study of galaxy evolution. To integrate the advantages of unsupervised learning without labels and supervised learning with high classification accuracy, this paper proposes a galaxy morphology classification model based on a momentum contrastive learning algorithm named Momentum Contrastive Learning Galaxy (MCL-Galaxy), which mainly includes two parts (i) pre-training of the model, where the ResNet_50 backbone network acts as an encoder to learn the galaxy morphology image features, which are stored in the queue and their consistency is ensured by using the momentum contrastive learning algorithm; and (ii) transfer learning, where Mahalanobis distance can assist in improving classification accuracy in downstream tasks where both encoder and queue are transferred. To evaluate the performance of MCL-Galaxy, we use the data set of the Galaxy Zoo challenge project on Kaggle for comparative testing. The experimental results show that the classification accuracy of MCL-Galaxy can reach 90.12%, which is 8.12% higher than the unsupervised state-of-the-art results. Although it is 3.1% lower than the advanced supervised method, it has the advantage of no label and can achieve a higher accuracy rate at the first epoch of classification iteration. This suggests that the gap between unsupervised and supervised representation learning in the field of Galaxy Morphologies classification tasks is well bridged.\",\"PeriodicalId\":20820,\"journal\":{\"name\":\"Publications of the Astronomical Society of the Pacific\",\"volume\":\"50 1\",\"pages\":\"0\"},\"PeriodicalIF\":3.3000,\"publicationDate\":\"2023-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Publications of the Astronomical Society of the Pacific\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1088/1538-3873/acf8f7\",\"RegionNum\":3,\"RegionCategory\":\"物理与天体物理\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ASTRONOMY & ASTROPHYSICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Publications of the Astronomical Society of the Pacific","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1088/1538-3873/acf8f7","RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ASTRONOMY & ASTROPHYSICS","Score":null,"Total":0}
引用次数: 0

摘要

星系形态的分类在天体物理学中占有重要地位,为星系演化的研究提供了很大的帮助。为了融合无标签无监督学习和有监督学习分类精度高的优点,本文提出了一种基于动量对比学习算法的星系形态分类模型,命名为动量对比学习星系(mcll - galaxy),该模型主要包括两个部分:一是模型的预训练,其中ResNet_50骨干网络作为编码器学习星系形态图像特征;并利用动量对比学习算法保证其一致性;(ii)迁移学习,其中马氏距离可以帮助提高下游任务的分类精度,其中编码器和队列都被转移。为了评估MCL-Galaxy的性能,我们使用Kaggle上Galaxy Zoo挑战项目的数据集进行对比测试。实验结果表明,MCL-Galaxy的分类准确率达到90.12%,比无监督状态下的分类准确率提高了8.12%。虽然比先进的监督方法低3.1%,但它具有无标签的优点,在分类迭代的第一个历元可以达到更高的准确率。这表明在星系形态分类任务领域中,无监督和有监督表示学习之间的差距已经很好地弥合了。
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
A Galaxy Morphology Classification Model Based on Momentum Contrastive Learning
Abstract The taxonomy of galaxy morphology plays an important role in astrophysics and provides great help for the study of galaxy evolution. To integrate the advantages of unsupervised learning without labels and supervised learning with high classification accuracy, this paper proposes a galaxy morphology classification model based on a momentum contrastive learning algorithm named Momentum Contrastive Learning Galaxy (MCL-Galaxy), which mainly includes two parts (i) pre-training of the model, where the ResNet_50 backbone network acts as an encoder to learn the galaxy morphology image features, which are stored in the queue and their consistency is ensured by using the momentum contrastive learning algorithm; and (ii) transfer learning, where Mahalanobis distance can assist in improving classification accuracy in downstream tasks where both encoder and queue are transferred. To evaluate the performance of MCL-Galaxy, we use the data set of the Galaxy Zoo challenge project on Kaggle for comparative testing. The experimental results show that the classification accuracy of MCL-Galaxy can reach 90.12%, which is 8.12% higher than the unsupervised state-of-the-art results. Although it is 3.1% lower than the advanced supervised method, it has the advantage of no label and can achieve a higher accuracy rate at the first epoch of classification iteration. This suggests that the gap between unsupervised and supervised representation learning in the field of Galaxy Morphologies classification tasks is well bridged.
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
Publications of the Astronomical Society of the Pacific
Publications of the Astronomical Society of the Pacific 地学天文-天文与天体物理
CiteScore
6.70
自引率
5.70%
发文量
103
审稿时长
4-8 weeks
期刊介绍: The Publications of the Astronomical Society of the Pacific (PASP), the technical journal of the Astronomical Society of the Pacific (ASP), has been published regularly since 1889, and is an integral part of the ASP''s mission to advance the science of astronomy and disseminate astronomical information. The journal provides an outlet for astronomical results of a scientific nature and serves to keep readers in touch with current astronomical research. It contains refereed research and instrumentation articles, invited and contributed reviews, tutorials, and dissertation summaries.
期刊最新文献
The Valuable Long-period Cluster Cepheid KQ Scorpii and other Calibration Candidates A New Parameterization for Finding Solutions for Microlensing Exoplanet Light Curves Multi-amplifier Sensing Charge-coupled Devices for Next Generation Spectroscopy Ejecta Masses in Type Ia Supernovae—Implications for the Progenitor and the Explosion Scenario* * Based in part on observations obtained with the Hobby-Eberly Telescope (HET), which is a joint project of the University of Texas at Austin, the Pennsylvania State University, Ludwig-Maximillians-Universitaet Muenchen, and Georg-August Universitaet Goettingen. The HET is named in honor of its principal benefactors, William P. Hobby and Robert E. Eberly. Physical Properties of Embedded Clusters in ATLASGAL Clumps with H ii Regions
×
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