Semantic Feature Learning for Heterogeneous Multitask Classification via Non-Negative Matrix Factorization

doi:10.1109/TCYB.2017.2732818

CORC > 计算技术研究所 > 中国科学院计算技术研究所 > 中国科学院计算技术研究所期刊论文 > 英文

	Semantic Feature Learning for Heterogeneous Multitask Classification via Non-Negative Matrix Factorization
	Zhuang, Fuzhen 2,3; Li, Xuebing 4,5; Jin, Xin 1; Zhang, Dapeng 5; Qiu, Lirong 6; He, Qing 2,3
刊名	IEEE TRANSACTIONS ON CYBERNETICS
	2018-08-01
卷号	48 期号:8 页码:2284-2293
关键词	Heterogeneous features multitask learning (MTL) non-negative matrix factorization (NMF) semantic feature learning
ISSN号	2168-2267
DOI	10.1109/TCYB.2017.2732818
英文摘要	Multitask learning (MTL) aims to learn multiple related tasks simultaneously instead of separately to improve the generalization performance of each task. Most existing MTL methods assumed that the multiple tasks to be learned have the same feature representation. However, this assumption may not hold for many real-world applications. In this paper, we study the problem of MTL with heterogeneous features for each task. To address this problem, we first construct an integrated graph of a set of bipartite graphs to build a connection among different tasks. We then propose a non-negative matrix factorization-based multitask method (MTNMF) to learn a common semantic feature space underlying different heterogeneous feature spaces of each task. Moreover, an improved version of MTNMF (IMTNMF) is proposed, in which we do not need to construct the correlation matrix between input features and class labels, avoiding the information loss. Finally, based on the common semantic features and original heterogeneous features, we model the heterogenous MTL problem as a multitask multiview learning (MTMVL) problem. In this way, a number of existing MTMVL methods can be applied to solve the problem effectively. Extensive experiments on three real-world problems demonstrate the effectiveness of our proposed methods, and the improved version IMTNMF can gain about 2% average accuracy improvement compared with MTNMF.
资助项目	National Natural Science Foundation of China[61473273] ; National Natural Science Foundation of China[91546122] ; National Natural Science Foundation of China[61573335] ; National Natural Science Foundation of China[6177021035] ; Guangdong Provincial Science and Technology plan projects[2015 B010109005] ; 2015 Microsoft Research Asia Collaborative Research Program ; Youth Innovation Promotion Association CAS[2017146]
WOS研究方向	Automation & Control Systems ; Computer Science
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000439363600006
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/4575]
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Zhuang, Fuzhen; Qiu, Lirong
作者单位	1.Huawei Technol Co Ltd, Shenzhen, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China 3.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 4.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China 5.Yanshan Univ, Coll Informat Sci & Engn, Qinhuangdao, Peoples R China 6.Minzu Univ China, Sch Informat Engn, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Zhuang, Fuzhen,Li, Xuebing,Jin, Xin,et al. Semantic Feature Learning for Heterogeneous Multitask Classification via Non-Negative Matrix Factorization[J]. IEEE TRANSACTIONS ON CYBERNETICS,2018,48(8):2284-2293.
APA	Zhuang, Fuzhen,Li, Xuebing,Jin, Xin,Zhang, Dapeng,Qiu, Lirong,&He, Qing.(2018).Semantic Feature Learning for Heterogeneous Multitask Classification via Non-Negative Matrix Factorization.IEEE TRANSACTIONS ON CYBERNETICS,48(8),2284-2293.
MLA	Zhuang, Fuzhen,et al."Semantic Feature Learning for Heterogeneous Multitask Classification via Non-Negative Matrix Factorization".IEEE TRANSACTIONS ON CYBERNETICS 48.8(2018):2284-2293.