Type
ArticleKAUST Department
Computational Bioscience Research Center (CBRC)Computer Science Program
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Date
2015-02-12Online Publication Date
2015-02-12Print Publication Date
2015-07Permanent link to this record
http://hdl.handle.net/10754/577311
Metadata
Show full item recordAbstract
In this paper we investigate the usage of regularized correntropy framework for learning of classifiers from noisy labels. The class label predictors learned by minimizing transitional loss functions are sensitive to the noisy and outlying labels of training samples, because the transitional loss functions are equally applied to all the samples. To solve this problem, we propose to learn the class label predictors by maximizing the correntropy between the predicted labels and the true labels of the training samples, under the regularized Maximum Correntropy Criteria (MCC) framework. Moreover, we regularize the predictor parameter to control the complexity of the predictor. The learning problem is formulated by an objective function considering the parameter regularization and MCC simultaneously. By optimizing the objective function alternately, we develop a novel predictor learning algorithm. The experiments on two challenging pattern classification tasks show that it significantly outperforms the machines with transitional loss functions.Citation
Regularized maximum correntropy machine 2015, 160:85 NeurocomputingPublisher
Elsevier BVJournal
NeurocomputingarXiv
1501.04282Additional Links
http://linkinghub.elsevier.com/retrieve/pii/S0925231215001150ae974a485f413a2113503eed53cd6c53
10.1016/j.neucom.2014.09.080