基于双重局部保持的不完整多视角嵌入学习方法

doi:10.19678/j.issn.1000-3428.0058129

摘要/Abstract

摘要： 现有的多视角降维方法多数假设数据是完整的，但该假设在实际应用中难以实现。为解决不完整多视角数据降维问题，提出一种新的不完整多视角嵌入学习方法。基于多视角数据的一致性与同一视角下样本间的线性相关性学习一组重构系数，对缺失样本进行线性重构，通过学习所有视角的公共低维嵌入，保持原始空间的局部几何结构。在此基础上，设计一种惩罚参数来度量重构样本的可靠度，从而权衡缺失样本对学习结果的负面影响。实验结果表明，该方法在Yale、ORL和COIL-20数据集上NMI值分别达到65.63%、73.23%和78.27%，较MVL-IV算法分别提升8.37%、16.71%和20.24%。

关键词: 多视角学习, 不完整数据, 降维, 嵌入学习, 局部保持

Abstract: Most of the existing multi-view dimensionality reduction methods assume that the data is complete, but it is unrealistic for practical applications.In order to solve the problems in the dimensionality reduction of incomplete multi-view data, this paper proposes a new method for incomplete multi-view embedded learning.Based on the consistency of multi-view data and the linear correlation between samples under the same view, a set of reconstruction coefficients is learnt for the reconstruction of the missing samples.Then the common low-dimensional embedding of all views is learnt to maintain the local geometric structure of the original space.Additionally, a penalty parameter is designed to measure the reliability of the reconstructed samples, so as to balance the negative influence of missing samples on the learning results.Experimental results show that the NMI value of the proposed method reaches 65.63%, 73.23% and 78.27% on the datasets of Yale, ORL and COIL-20, increased by 8.37%, 16.71% and 20.24% compared with the MVL-IV algorithm.

Key words: multi-view learning, incomplete data, dimensionality reduction, embedded learning, locality preserving

中图分类号:

TP18

刘彦雯, 张金鑫, 张宏杰, 经玲. 基于双重局部保持的不完整多视角嵌入学习方法[J]. 计算机工程, 2021, 47(6): 115-122,141.

LIU Yanwen, ZHANG Jinxin, ZHANG Hongjie, JING Ling. Incomplete Multi-view Embedded Learning Method Based on Double Locality Preserving[J]. Computer Engineering, 2021, 47(6): 115-122,141.

http://www.ecice06.com/CN/Y2021/V47/I6/115

图/表 10

20210618183226

20210618183230

20210618183233

20210618183237

20210618183240

20210618183244

20210618183247

20210618183251

20210618183255

20210618183259

参考文献

[1] TANG Jingjing,TIAN Yingjie.Overview of multi-view learning[J].Mathematical Modeling and Its application,2017,6(3):1-15.(in Chinese)唐静静,田英杰.多视角学习综述[J].数学建模及其应用,2017,6(3):1-15.
[2] ZONG Linlin,ZHANG Xianchao,ZHAO Long,et al.Multi-view clustering via multi-manifold regularized non-negative matrix factorization[J].Neural Networks,2017,88:74-89.
[3] LIU Jialu,WANG Chi,GAO Jing,et al.Multi-view clustering via joint nonnegative matrix factorization[C]//Proceedings of 2013 SIAM International Conference on Data Mining.Austin,USA:[s.n.],2013:252-260.
[4] SUN Shiliang.A survey of multi-view machine learning[J].Neural Computing and Applications,2013,23(7):2031-2038.
[5] COLLINS M,SINGER Y.Unsupervised models for named entity classification[C]//Proceedings of 1999 Joint SIGDAT Conference on Empirical Methods in Natural Language Processing and Very Large Corpora.Florham,USA:[s.n.],1999:100-110.
[6] GHANI R.Combining labeled and unlabeled data for text classification with a large number of categories[C]//Proceedings of International Conference on Data Mining.San Jose,USA:IEEE Press,2001:597-598.
[7] MUSLEA I,MINTON S,KNOBLOCK C,et al.Active learning with multiple views[J].Journal of Artificial Intelligence Research,2006,27(1):203-233.
[8] BLUM A,MITCHELL T M.Combining labeled and unlabeled data with co-training[C]//Proceedings of the 11th Annual Conference on Learning Theory.Washington D.C.,USA:IEEE Press,1998:92-100.
[9] HECKMANN M,BERTHOMMIER F,KROSCHEL K.Noise adaptive stream weighting in audio-visual speech recognition[J].EURASIP Journal on Advances in Signal Processing,2002(11):1-14.
[10] WU Y,CHANG E Y,CHANG K C C,et al.Optimal multimodal fusion for multimedia data analysis[C]//Proceedings of the 12th Annual ACM International Conference on Multimedia.New York,USA:ACM Press:2004:572-579.
[11] WANG Yang,ZHANG Wenjie,WU Lin,et al.Unsupervised metric fusion over multiview data by graph random walk-based cross-view diffusion[J].IEEE Transactions on Neural Networks and Learning Systems,2017,28(1):57-70.
[12] INDYK P,MOTWANI R.Approximate nearest neighbors:towards removing the curse of dimensionality[C]//Proceedings of the 30th Annual ACM Symposium on Theory of Computing.New York,USA:ACM Press,1998:604-613.
[13] JOLLIFFE I T.Principal Component Analysis[M].Berlin,Germany:Springer,2011.
[14] BELKIN M,NIYOGI P.Laplacian eigenmaps for dimensionality reduction and data representation[J].Neural computation,2003,15(6):1373-1396.
[15] DER M L V,HINTON G E.Visualizing data using t-SNE[J].Journal of Machine Learning Research,2008,31(5):2579-2605.
[16] MOORE B A,MCLACHLAN G J.Discriminant analysis and statistical pattern recognition[J].Biometrical Journal,1993,35(7):784-784.
[17] XIA Tian,TAO Dacheng,MEI Tao,et al.Multiview spectral embedding[J].IEEE Transactions on Systems,Man and Cybernetics,2010,40(6):1438-1446.
[18] KAMALIKA C,SHAM M K.Multi-view clustering via canonical correlation analysis[C]//Proceedings of International Conference on Machine Learnig.Montreal,Canada:[s.n.],2009:129-136.
[19] TAEKYUN K,JOSEF K,ROBERTO C.Discriminative learning and recognition of image set classes using canonical correlations[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(6):1005-1018.
[20] HU Menglei,CHEN Songcan.Doubly aligned incomplete multi-view clustering[C]//Proceedings of IEEE International Joint Conference on Artificial Intelligence.Washington D.C.,USA:IEEE Press,2018:2262-2268.
[21] LI Shaoyuan,JIANG Yuan,ZHOU Zhihua.Partial multi-view clustering[C]//Proceedings of AAAI Conference on Artificial Intelligence.[S.1.]:AAAI Press,2014:1968-1974.
[22] CHRISTOPH H L,OLIVER K.Weakly-paired maximum covariance analysis for multimodal dimensionality reduction and transfer learning[C]//Proceedings of European Conference on Computer Vision.Berlin,Germany:Springer,2010:566-579.
[23] CHEN Xiaohong,CHEN Songcan,XUE Hui,et al.A unified dimensionality reduction framework for semi-paired and semi-supervised multi-view data[J].Pattern Recognition,2012,45(5):2005-2018.
[24] KIMURA A,KAMEOKA H,SUGIYAMA M,et al.SemiCCA:efficient semi-supervised learning of canonical correlations[C]//Proceedings of the 20th International Conference on Pattern Recognition.Washington D.C.,USA:IEEE Press,2010:2933-2936.
[25] HOTELING H.Relations between two sets of variates[J].Biometrika,1936,28(4):321-377.
[26] HARDOON D R,SZEDMAK S,SHAWETYLOR J,et al.Canonical correlation analysis:an over view with application to learning methods[J].Neural Computation,2004,16(12):2639-2664.
[27] TRIVEDI A,RAI P,DUVALL S.Multi-view clustering with incomplete views[C]//Proceedings of Workshop on Machine Learning for Social Computing.Whistler,Canada:[s.n.],2010:656-579.
[28] XU Chang,TAO Dacheng,XU Chao.Multi-view learning with incomplete views[J].IEEE Transactions on Image Processing,2015,24(12):5812-5825.
[29] ARGYRIOU A,EVGENIOU T,PONYIL M.Multi-task feature learning[C]//Proceedings of Advances in Neural Information Processing Systems.Washington D.C.,USA:IEEE Press,2007:19-41.
[30] WANG Xin,BI Jinbo,YU Shipeng,et al.On multiplicative multitask feature learning[C]//Proceedings of Advances in Neural Information Processing Systems.Washington D.C.,USA:IEEE Press,2014:2411-2419.
[31] TAO Hong,HOU Chenping,YI Dongyun,et al.Joint embedding learning and low-rank approximation:a framework for incomplete multi-view learning[J].IEEE Transactions on Cybernetics,2019,65(5):1-14.
[32] YANG Wanqi,SHI Yinghuan,GAO Yang,et al.Incomplete-data oriented multiview dimension reduction via sparse low-rank representation[J].IEEE Transactions on Neural Networks and Learning Systems,2018,29(12):6276-6291.
[33] CHEN Ning,ZHU Jun,SUN Fuchun,et al.Large-margin predictive latent subspace learning for multi-view data analysis[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(12):2365-2378.
[34] MARTHA W,ZHANG X,DALE S,et al.Convex multi-view subspace learning[C]//Proceedings of NIPS'12.Cambridge,USA:MIT Press,2012:1673-1681.
[35] SHARMA A,KUMAR A,DAUME H,et al.Generalized multi-view analysis:a discriminative latent space[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.[S.1.]:IEEE Computer Society,2012:2160-2167.
[36] ZHANG Lei,ZHAO Yao,ZHU Zhenfeng,et al.Multi-view missing data completion[J].IEEE Transactions on Knowledge and Data Engineering,2018,30(7):1296-1309.
[37] SAM T R,LAWRENCE K S.Nonlinear dimension reduction by locally linear embedding[J].Science,2000,290:2323-2326.
[38] HE X F,PARTHA N.Locality preserving projections[C]//Proceedings of NIPS'02.Cambridge,USA:MIT Press,2002:153-160.
[39] YIN Qiyue,WU Shu,WANG Liang.Multiview clustering via unified and view-specific embeddings learning[J].IEEE Transactions on Neural Networks and Learning Systems,2018,29(11):5541-5553.

选择文件类型/文献管理软件名称

选择包含的内容