Titlebook: Robust Recognition via Information Theoretic Learning; Ran He,Baogang Hu,Liang Wang Book 2014 The Author(s) 2014 Face recognition.informat

減震

,, Regularized Correntropy,Sparse signal representation arises in application of compressed sensing and has been considered as a significant technique in computer vision and machine learning [27, 65, 154]. Based on the ..-.. equivalence theory [18, 39], the solution of an ..-minimization problem is equal to that of an .. minimization problem under certain conditions.

correspondent

Ran He,Baogang Hu,Liang WangIncludes supplementary material:

跳動(dòng)

Robust Recognition via Information Theoretic Learning978-3-319-07416-0Series ISSN 2191-5768 Series E-ISSN 2191-5776

constitute

Introduction,ived from the statistical definition of a breakdown point [49, 106], is the ability of an algorithm that tolerates a large amount of outliers. Therefore, a robust method should be effective enough to reject outliers in images and perform classification only on uncorrupted pixels. In the past decades

artifice

M-Estimators and Half-Quadratic Minimization,binations of order statistics), R-estimator (estimator based on rank transformation) [77], RM estimator (repeated median) [141], and LMS estimator (estimator using the least median of squares) [133]. When information theoretic learning is applied to robust statistics, the Gaussian kernel in entropy

Palpate

DUST

CANDY

CRANK

overreach

Correntropy and Linear Representation,samples are available. However, in practice, only a small number of samples are available for an object class. Hence linear representation methods are developed to generalize the representational capacity of available samples.