Maximum Margin Multi-Dimensional Classification

doi:10.1109/TNNLS.2021.3084373

	Maximum Margin Multi-Dimensional Classification
	Jia, Bin-Bin 2,3; Zhang, Min-Ling 1,3
刊名	IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
	2021
关键词	Covariance matrices Training Predictive models Optimization Task analysis Semantics Matrix decomposition Class dependencies machine learning maximum margin multi-dimensional classification (MDC)
ISSN号	2162-237X
DOI	10.1109/TNNLS.2021.3084373
英文摘要	Multi-dimensional classification (MDC) assumes heterogeneous class spaces for each example, where class variables from different class spaces characterize semantics of the example along different dimensions. The heterogeneity of class spaces leads to incomparability of the modeling outputs from different class spaces, which is the major difficulty in designing MDC approaches. In this article, we make a first attempt toward adapting maximum margin techniques for MDC problem and a novel approach named (MMDC)-M-3 is proposed. Specifically, (MMDC)-M-3 maximizes the margins between each pair of class labels with respect to individual class variable while modeling relationship across class variables (as well as class labels within individual class variable) via covariance regularization. The resulting formulation admits convex objective function with nonlinear constraints, which can be solved via alternating optimization with quadratic programming (QP) or closed-form solution in either alternating step. Comparative studies on the most comprehensive real-world MDC datasets to date are conducted and it is shown that (MMDC)-M-3 achieves highly competitive performance against state-of-the-art MDC approaches.
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000732406600001
内容类型	期刊论文
源URL	[http://ir.lut.edu.cn/handle/2XXMBERH/155016]
专题	电气工程与信息工程学院
作者单位	1.Southeast Univ, Minist Educ China, Key Lab Comp Network & Informat Integrat, Nanjing, Jiangsu, Peoples R China 2.Lanzhou Univ Technol, Coll Elect & Informat Engn, Lanzhou 730050, Peoples R China; 3.Southeast Univ, Sch Comp Sci & Engn, Nanjing 210096, Peoples R China;
推荐引用方式 GB/T 7714	Jia, Bin-Bin,Zhang, Min-Ling. Maximum Margin Multi-Dimensional Classification[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,2021.
APA	Jia, Bin-Bin,&Zhang, Min-Ling.(2021).Maximum Margin Multi-Dimensional Classification.IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS.
MLA	Jia, Bin-Bin,et al."Maximum Margin Multi-Dimensional Classification".IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2021).