设为首页收藏本站

VALSE

VALSE 首页 活动通知 好文作者面授招 查看内容

20160810-26罗勇:Tensor Canonical Correlation Analysis and Its applications

2016-8-6 08:24| 发布者: 程一-计算所| 查看: 6091| 评论: 0

摘要: 报告嘉宾:罗勇(南洋理工大学)报告时间: 2016年8月10日(星期三)晚20:00(北京时间)主持人: 王楠楠(西安电子科技大学)报告题目:Tensor Canonical Correlation Analysis and Its applications文章信息:Ten ...

  • 报告嘉宾:罗勇(南洋理工大学)
  • 报告时间: 2016年8月10日(星期三)晚20:00(北京时间)
  • 主持人: 王楠楠(西安电子科技大学)
  • 报告题目:Tensor Canonical Correlation Analysis and Its applications

  • 文章信息:
    1. Tensor Canonical Correlation Analysis for Multi-view Dimension Reduction, Yong Luo, Dacheng Tao, Kotagiri Ramamohanarao, Chao Xu, and Yonggang Wen, IEEE Transactions on Knowledge and Data Engineering (T-KDE), vol. 27, no. 11, pp. 3111-3124, 2015.
    2. On Combining Side Information and Unlabeled Data for Heterogeneous Multi-task Metric Learning, Yong Luo, Yonggang Wen and Dacheng Tao, International Joint Conference on Artificial Intelligence (IJCAI), pp. 1809-1815, 2016.

  • 报告摘要:
  • Canonical correlation analysis (CCA) has proven an effective tool for two-view dimension reduction due to its profound theoretical foundation and success in practical applications. In respect of multi-view learning, however, it is limited by its capability of only handling data represented by two-view features, while in many real-world applications, the number of views is frequently many more. Although the ad hoc way of simultaneously exploring all possible pairs of features can numerically deal with multi-view data, it ignores the high order statistics (correlation information) which can only be discovered by simultaneously exploring all features. Therefore, We develop tensor CCA (TCCA) which straightforwardly yet naturally generalizes CCA to handle the data of an arbitrary number of views by analyzing the covariance tensor of the different views. As a consequence, the high-order correlation information contained in the different views is explored and thus a reliable common subspace shared by all features can be obtained. In addition to multi-view dimension reduction, we also apply TCCA to heterogeneous transfer metric learning by combining side information and unlabeled data. By maximizing the high-order correlation between all domains, much more correlation information can be encoded in the learned metrics, and hopefully better performance can be achieved.

  • 报告人简介:
  • Yong Luo received the B.E. degree in Computer Science from the Northwestern Polytechnical University, Xi'an, China, in 2009, and the D.Sc. degree in the School of Electronics Engineering and Computer Science, Peking University, Beijing, China, in 2014. He was a Research Fellow with the School of Computer Science and Engineering, Nanyang Technological University, Singapore. He was also a visiting student in the School of Computer Engineering, Nanyang Technological University, and the Faculty of Engineering and Information Technology, University of Technology Sydney. His research interests are primarily on machine learning with applications on image classification and annotation. He has authored several scientific articles at top venues including IEEE T-NNLS, IEEE T-IP, IEEE T-KDE, AAAI, and IJCAI.

最新评论

Archiver|手机版|小黑屋|Vision And Learning SEminar    

GMT+8, 2020-9-27 00:30 , Processed in 0.028665 second(s), 17 queries .

Powered by Discuz! X3.2

© 2001-2013 Comsenz Inc.

返回顶部