基于GAN的聚类解生成与扩散融合

IF 3.2 Q2 AUTOMATION & CONTROL SYSTEMS
Wenming Cao, Zhiwen Yu, H. Wong
{"title":"基于GAN的聚类解生成与扩散融合","authors":"Wenming Cao, Zhiwen Yu, H. Wong","doi":"10.1080/21642583.2021.2024100","DOIUrl":null,"url":null,"abstract":"In this paper, we propose a framework to generate diverse clustering solutions and conduct solution retrieval to improve performance. Specifically, we first project unlabelled data from multiple domains into a shared space while preserving the respective semantics. This space allows that representations of samples in a hard domain are recovered by a linear combination of those of others in the easy domains. Meanwhile, a clustering algorithm is adopted to provide pseudo labels for a conditional generative adversarial network to synthesize representations that in turn promote the learning of the above space. Second, we conduct the joint learning of feature projection and partition matrices on batches of representations, where the former ones are considered as clustering solutions and input into another generative adversarial network to generate more solutions. Third, we utilize the fusion of diffusion to effectively retrieve and extract the knowledge in multiple solutions to obtain the final clustering. We perform comparative experiments against other methods on multiple benchmark data sets. Experimental results demonstrate the effectiveness and superiority of our proposed method.","PeriodicalId":46282,"journal":{"name":"Systems Science & Control Engineering","volume":null,"pages":null},"PeriodicalIF":3.2000,"publicationDate":"2022-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"GAN-based clustering solution generation and fusion of diffusion\",\"authors\":\"Wenming Cao, Zhiwen Yu, H. Wong\",\"doi\":\"10.1080/21642583.2021.2024100\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this paper, we propose a framework to generate diverse clustering solutions and conduct solution retrieval to improve performance. Specifically, we first project unlabelled data from multiple domains into a shared space while preserving the respective semantics. This space allows that representations of samples in a hard domain are recovered by a linear combination of those of others in the easy domains. Meanwhile, a clustering algorithm is adopted to provide pseudo labels for a conditional generative adversarial network to synthesize representations that in turn promote the learning of the above space. Second, we conduct the joint learning of feature projection and partition matrices on batches of representations, where the former ones are considered as clustering solutions and input into another generative adversarial network to generate more solutions. Third, we utilize the fusion of diffusion to effectively retrieve and extract the knowledge in multiple solutions to obtain the final clustering. We perform comparative experiments against other methods on multiple benchmark data sets. Experimental results demonstrate the effectiveness and superiority of our proposed method.\",\"PeriodicalId\":46282,\"journal\":{\"name\":\"Systems Science & Control Engineering\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.2000,\"publicationDate\":\"2022-01-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Systems Science & Control Engineering\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1080/21642583.2021.2024100\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"AUTOMATION & CONTROL SYSTEMS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Systems Science & Control Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1080/21642583.2021.2024100","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"AUTOMATION & CONTROL SYSTEMS","Score":null,"Total":0}
引用次数: 1

摘要

在本文中,我们提出了一个框架来生成不同的集群解决方案,并进行解决方案检索以提高性能。具体来说,我们首先将来自多个域的未标记数据投影到共享空间中,同时保留各自的语义。该空间允许通过易域中其他样本的线性组合来恢复硬域中样本的表示。同时,采用聚类算法为条件生成对抗性网络提供伪标签,以合成表示,从而促进对上述空间的学习。其次,我们对一批表示进行特征投影和划分矩阵的联合学习,其中前者被视为聚类解,并输入到另一个生成对抗性网络中以生成更多的解。第三,我们利用扩散的融合来有效地检索和提取多个解中的知识,以获得最终的聚类。我们在多个基准数据集上与其他方法进行比较实验。实验结果证明了该方法的有效性和优越性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
GAN-based clustering solution generation and fusion of diffusion
In this paper, we propose a framework to generate diverse clustering solutions and conduct solution retrieval to improve performance. Specifically, we first project unlabelled data from multiple domains into a shared space while preserving the respective semantics. This space allows that representations of samples in a hard domain are recovered by a linear combination of those of others in the easy domains. Meanwhile, a clustering algorithm is adopted to provide pseudo labels for a conditional generative adversarial network to synthesize representations that in turn promote the learning of the above space. Second, we conduct the joint learning of feature projection and partition matrices on batches of representations, where the former ones are considered as clustering solutions and input into another generative adversarial network to generate more solutions. Third, we utilize the fusion of diffusion to effectively retrieve and extract the knowledge in multiple solutions to obtain the final clustering. We perform comparative experiments against other methods on multiple benchmark data sets. Experimental results demonstrate the effectiveness and superiority of our 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
Systems Science & Control Engineering
Systems Science & Control Engineering AUTOMATION & CONTROL SYSTEMS-
CiteScore
9.50
自引率
2.40%
发文量
70
审稿时长
29 weeks
期刊介绍: Systems Science & Control Engineering is a world-leading fully open access journal covering all areas of theoretical and applied systems science and control engineering. The journal encourages the submission of original articles, reviews and short communications in areas including, but not limited to: · artificial intelligence · complex systems · complex networks · control theory · control applications · cybernetics · dynamical systems theory · operations research · systems biology · systems dynamics · systems ecology · systems engineering · systems psychology · systems theory
×
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