[1]郭蔚颖,房小兆,吴宝昌*,等.基于低秩交叉重构的领域自适应算法[J].江西师范大学学报(自然科学版),2021,(04):390-397.[doi:10.16357/j.cnki.issn1000-5862.2021.04.11]
 GUO Weiying,FANG Xiaozhao,WU Baochang*,et al.The Low-Rank Constraint-Based Cross Reconstruction for Domain Adaptation[J].Journal of Jiangxi Normal University:Natural Science Edition,2021,(04):390-397.[doi:10.16357/j.cnki.issn1000-5862.2021.04.11]
点击复制

基于低秩交叉重构的领域自适应算法()
分享到:

《江西师范大学学报》(自然科学版)[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2021年04期
页码:
390-397
栏目:
信息科学与技术
出版日期:
2021-08-10

文章信息/Info

Title:
The Low-Rank Constraint-Based Cross Reconstruction for Domain Adaptation
文章编号:
1000-5862(2021)04-0390-08
作者:
郭蔚颖1房小兆2吴宝昌3*滕少华1
1.广东工业大学计算机学院,广东 广州 510006; 2.广东工业大学自动化学院,广东 广州 510006; 3.广东金融学院公共管理学院,广东 广州 510520
Author(s):
GUO Weiying1FANG Xiaozhao2WU Baochang3*TENG Shaohua1
1.School of Computers,Guangdong University of Technology,Guangzhou Guangdong 510006,China; 2.School of Automation,Guangdong University of Technology,Guangzhou Guangdong 510006,China; 3.School of Public Administration,Guangdong University of Finance,Guangzhou Guangdong 510520,China
关键词:
领域自适应 交叉重构 低秩约束 跨域识别
Keywords:
domain adaptation cross reconstruction low-rank constraint cross-domain recognition
分类号:
TP 391
DOI:
10.16357/j.cnki.issn1000-5862.2021.04.11
文献标志码:
A
摘要:
为了解决现有领域在自适应方法中忽略了整个数据域内部结构的信息和源域与目标域之间的差异问题,提出了一种新的基于低秩交叉重构的领域自适应方法.通过对源域和目标域的交叉重构来构造新的源域与目标域,使得同类数据相互交织,缩短了同类数据之间的距离; 通过对重构矩阵施加低秩约束,将2个域的同类数据对齐,以此来充分挖掘源域和目标域同类数据之间的内在结构信息,并利用该结构信息来学习分类器,从而取得更好的跨域识别效果.在5个公开数据集上的实验结果表明:该方法具有较高的跨域识别准确率.
Abstract:
In order to solve these problems,a new low-rank constraint-based cross reconstruction(LRCR)for domain adaptation is proposed.Through the cross reconstruction of the source domain and the target domain,the new source domain and the target domain are constructed so that the same label data are interweaved with each other.By applying the low-rank constraint,the same label data in the two domains are aligned to fully mine the internal structure information between the same label data in the source domain and the target domain.The structure information is used to learn the classifier to achieve better cross-domain recognition accuracy.The experimental results on five public datasets show that LRCR has high cross-domain recognition accuracy.

参考文献/References:

[1] Weiss K,Khoshgoftaar T M,Wang Dingding.A survey of transfer learning [J].Journal of Big Data,2016,3(1):1-40.
[2] Dai Wenyuan,Yang Qiang,Xue Guirong,et al.Boosting for transfer learning [EB/OL].[2020-11-23].https://doi.org/10.1145/1273496.1273521.
[3] Pan Sinno Jialin,Yang Qiang.A survey on transfer learning [J].IEEE Transactions on Knowledge and Data Engineering,2010,22(10):1345-1359.
[4] 庄福振,罗平,何清,等.迁移学习研究进展 [J].软件学报,2015,26(1):26-39.
[5] Lin Zhouchen,Zhang Chao,Zhou Pan.Integrated low-rank-based discriminative feature learning for recognition [J].IEEE Transactions on Neural Networks and Learning Systems,2016,27(5):1080-1093.
[6] Liu Guangcan,Lin Zhouchen,Yan Shuicheng,et al.Robust recovery of subspace structures by low-rank representation [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(1):171-184.
[7] Liu Guangcan,Yan Shuichen.Latent low-rank representation for subspace segmentation and feature extraction [EB/OL].[2020-06-17].https://dl.acm.org/doi/10.1109/ICCV.2011.6126422.
[8] Fang Xiaozhao,Han Na,Wu Jigang,et al.Approximate low-rank projection learning for feature extraction [J].IEEE Transactions on Neural Networks and Learning Systems,2018,29(11):5228-5241.
[9] 靳洋.基于PCA/ICA的图像特征提取算法研究 [D].西安:西安电子科技大学,2014.
[10] Hervé Abdi,Williams L J.Principal component analysis [J].Wiley Interdisciplinary Reviews Computational Statistics,2010,2(4):433-459.
[11] Pan Sinno Jialin,Tsang I W,Kwok J T,et al.Domain adaptation via transfer component analysis [J].IEEE Transactions on Neural Networks,2011,22(2):199-210.
[12] 屈磊,方怡,熊友玲,等.基于L2,1模和图正则化的低秩迁移子空间学习 [J].控制理论与应用,2018,35(12):1738-1749.
[13] Georghiades A S,Belhumeur P N,Kriegman D J.From few to many:illumination cone models for face recognition under variable lighting and pose [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,23(6):643-660.
[14] 吴越,曾向荣,周典乐,等.图像恢复中的稳健交替方向乘子法 [J].国防科技大学学报,2018,40(2):115-121.
[15] Tai Ying,Yang Jian,Zhang Yigong,et al.Face recognition with pose variations and misalignment via orthogonal procrustes regression [J].IEEE Transactions on Image Processing,2016,25(6):2673-2683.
[16] Fridman E.Descriptor discretized Lyapunov functional method [J].IFAC Proceedings Volumes,2006,39(9):638-643.
[17] Yang Junfeng,Yin Wotao,Zhang Yin,et al.A fast algorithm for edge-preserving variational multichannel image restoration [J].SIAM Journal on Imaging Sciences,2009,2(2):569-592.
[18] Rate C,Retrieval C.Columbia object image library(COIL-20)[EB/OL].[2020-06-17].https://www.igi-global.com/pdf.aspx?tid=104931&ptid=46172&ctid=17&t=index.
[19] Galleguillos C,Mcfee B,Belongie S,et al.From region similarity to category discovery [EB/OL].[2020-06-17].https://dl.acm.org/doi/10.1109/CVPR.2011.5995527.
[20] Fernando B,Tommasi T,Tuytelaars T.Joint cross-domain classification and subspace learning for unsupervised adaptation [J].Pattern Recognition Letters,2015,65(1):60-66.
[21] Liang Jian,He Ran,Sun Zhenan,et al.Aggregating randomized clustering-promoting invariant projections for domain adaptation [EB/OL].[2020-12-11].https://ieeexplore.ieee.org/document/8353356.
[22] Gong Boqing,Shi Yuan,Sha Fei,et al.Geodesic flow kernel for unsupervised domain adaptation [EB/OL].[2020-06-17].https://ieeexplore.ieee.org/document/6247911.
[23] Shao Ming,Kit D,Fu Yun.Generalized transfer subspace learning through low-rank constraint [J].International Journal of Computer Vision,2014,109(1/2):74-93.
[24] Yang Meng,Zhang Lei,Feng Xiangchu,et al.Sparse representation based fisher discrimination dictionary learning for image classification [J].International Journal of Computer Vision,2014,109(3):209-232.
[25] Zhang Jing,Li Wanqing,Ogunbona P.Joint geometrical and statistical alignment for visual domain adaptation [EB/OL].[2020-06-17].http://www.doc88.com/p-2374903824918.html.
[26] Zhu Fan,Shao Ling.Weakly-supervised cross-domain dictionary learning for visual recognition [J].International Journal of Computer Vision,2014,109(1/2):42-59.
[27] Wang Jindong,Feng Wenjie,Chen Yiqiang,et al.Visual domain adaptation with manifold embedded distribution alignment [EB/OL].[2020-06-17].https://arxiv.org/abs/1807.07258.
[28] Jia Zehui,Guo Ke,Cai Xingju.Convergence analysis of alternating direction method of multipliers for a class of separable convex programming [J].Abstract and Applied Analysis,2013(2):205-215.
[29] Hong Mingyi,Luo Zhijuan,Razaviyayn M.Convergence analysis of alternating direction method of multipliers for a family of nonconvex problems [EB/OL].[2020-06-17].https://arxiv.org/abs/1410.1390.
[30] Hong Mingyi,Luo Zhijuan.On the linear convergence of the alternating direction method of multipliers [J].Mathematical Programming,2012,162(1/2):1-29.

备注/Memo

备注/Memo:
收稿日期:2021-02-17
基金项目:国家自然科学基金(61772141,U1911401,61972102),广东省科技计划(2019B020208001,2019B110210002),广东省重点领域研发计划(2019B010121001,2019B010118001,2019B010119001)和广州市科技计划(201903010107)资助项目.
通信作者:吴宝昌(1978—),男,广东廉江人,讲师,主要从事电子政务、大数据与公共治理的研究.E-mail:8426008@qq
更新日期/Last Update: 2021-08-10