Representative Selection with Structured Sparsity

PATTERN RECOGNITION

卷:63

Article

Wang, Hongxing^1,2,3  Kawahara, Yoshinobu^4,5  Weng, Chaoqun³  Yuan, Junsong³ 

[1] Chongqing Univ, Minist Educ, Key Lab Dependable Serv Comp Cyber Phys Soc, Chongqing 400044, Peoples R China

[2] Chongqing Univ, Sch Software Engn, Chongqing 401331, Peoples R China

[3] Nanyang Technol Univ, Sch Elect & Elect Engn, Singapore 639798, Singapore

[4] Osaka Univ, Inst Sci & Ind Res, Osaka 5670047, Japan

[5] RIKEN, Ctr Adv Integrated Intelligence Res, Saitama 3510198, Japan

关键词: Representative selection;    Structured sparsity;    Diversity;   

DOI  :  10.1016/j.patcog.2016.10.014

来源: Elsevier

【 摘 要 】

We propose a novel formulation to find representatives in data samples via learning with structured sparsity. To find representatives with both diversity and representativeness, we formulate the problem as a structurally regularized learning where the objective function consists of a reconstruction error and three structured regularizers: (1) group sparsity regularizer, (2) diversity regularizer, and (3) locality-sensitivity regularizer. For the optimization of the objective, we propose an accelerated proximal gradient algorithm, combined with the proximal-Dykstra method and the calculation of parametric maximum flows. Experiments on image and video data validate the effectiveness of our method in finding exemplars with diversity and representativeness and demonstrate its robustness to outliers.

【 授权许可】

Free