Greedy Matrix Completion with Fitting Error and Rank Iterative Minimization  

作　　者：WANG Youhua ZHANG Yiming ZHANG Jianqiu HU Bo 

机构地区：[1]Key Laboratory of EMW Information (MoE) and the Department of Electronic Engineering, Fudan University [2]Science and Technology on Analog Integrated Circuit Laboratory [3]Global Academic Fellow in Statistics, New York University Abu Dhabi

出　　处：《Chinese Journal of Electronics》2017年第4期814-819,共6页电子学报（英文版）

基　　金：supported by the National Natural Science Foundation of China(No.61171127,No.61571131)

摘　　要：A novel matrix completion algorithm which iteratively minimizes the fitting error and the matrix rank is presented. Unlike conventional matrix completion algorithms, which usually require some relaxation technique to cope with the low rank constraints, the proposed algorithm does not require any such techniques, thus making the selection of the parameter q of the matrix qnorm(0 < q ≤ 1) or the regularization parameter unnecessary. Simulation results of the random generated data and Jester joke data set verify our algorithm's effectiveness and superiority over the reported algorithms in literature.A novel matrix completion algorithm which iteratively minimizes the fitting error and the matrix rank is presented. Unlike conventional matrix completion algorithms, which usually require some relaxation technique to cope with the low rank constraints, the proposed algorithm does not require any such techniques, thus making the selection of the parameter q of the matrix qnorm（0 〈 q ≤ 1） or the regularization parameter unnecessary. Simulation results of the random generated data and Jester joke data set verify our algorithm＇s effectiveness and superiority over the reported algorithms in literature.

关 键 词：Low-rankness Matrix completion Spar sity Ma jorization minimization Singular value decompo sition 

分 类 号：O241.6[理学—计算数学]