Regularized maximum correntropy machine

Jim Jing-Yan Wang, Yunji Wang, Bing-Yi Jing, Xin Gao

Research output: Contribution to journalArticlepeer-review

20 Scopus citations

Abstract

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.
Original languageEnglish (US)
Pages (from-to)85-92
Number of pages8
JournalNeurocomputing
Volume160
DOIs
StatePublished - Feb 12 2015

ASJC Scopus subject areas

  • Artificial Intelligence
  • Cognitive Neuroscience
  • Computer Science Applications

Fingerprint Dive into the research topics of 'Regularized maximum correntropy machine'. Together they form a unique fingerprint.

Cite this