Robust Volume Minimization-Based Matrix Factorization for Remote Sensing and Document Clustering
Publication in refereed journal


香港中文大学研究人员 ( 现职)

马荣健教授 (电子工程学系)

全文

数位物件识别号 (DOI) http://dx.doi.org/10.1109/TSP.2016.2602800

引用次数
Web of Sciencehttp://aims.cuhk.edu.hk/converis/portal/Publication/1WOS source URL

其它资讯

摘要This paper considers volume minimization (VolMin)-based structured matrix factorization. VolMin is a factorization criterion that decomposes a given data matrix into a basis matrix times a structured coefficient matrix via finding the minimum-volume simplex that encloses all the columns of the data matrix. Recent work showed that VolMin guarantees the identifiability of the factor matrices under mild conditions that are realistic in a wide variety of applications. This paper focuses on both theoretical and practical aspects of VolMin. On the theory side, exact equivalence of two independently developed sufficient conditions for VolMin identifiability is proven here, thereby providing a more comprehensive understanding of this aspect of VolMin. On the algorithm side, computational complexity and sensitivity to outliers are two key challenges associated with real-world applications of VolMin. These are addressed here via a new VolMin algorithm that handles volume regularization in a computationally simple way, and automatically detects and iteratively downweights outliers, simultaneously. Simulations and real-data experiments using a remotely sensed hyperspectral image and the Reuters document corpus are employed to showcase the effectiveness of the proposed algorithm.

着者Fu X, Huang KJ, Yang B, Ma WK, Sidiropoulos ND
期刊名称IEEE Transactions on Signal Processing
出版年份20http://aims.cuhk.edu.hk/converis/portal/Publication/16
月份http://aims.cuhk.edu.hk/converis/portal/Publication/12
日期http://aims.cuhk.edu.hk/converis/portal/Publication/1
卷号64
期次23
出版社IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
页次6254 - 6268
国际标準期刊号http://aims.cuhk.edu.hk/converis/portal/Publication/1053-587X
电子国际标準期刊号http://aims.cuhk.edu.hk/converis/portal/Publication/194http://aims.cuhk.edu.hk/converis/portal/Publication/1-0476
语言英式英语

关键词Document clustering; hyperspectral unmixing; identifiability; matrix factorization; robustness against outliers; simplex-volume minimization (VolMin)
Web of Science 学科类别Engineering; Engineering, Electrical & Electronic