DocumentCode :
3580076
Title :
Empirical survival error potential weighted least squares for binary pattern classification
Author :
Lei Sun ; Kar-Ann Toh ; Zhiping Lin ; Badong Chen
Author_Institution :
Sch. of Inf. & Electron., Beijing Inst. of Technol. Beijing, Beijing, China
fYear :
2014
Firstpage :
949
Lastpage :
952
Abstract :
A weighted least squares scheme based on an empirical survival error potential function is proposed in this paper. The empirical survival error potential function provides an error compensation scheme for noise distributions far from being Gaussian. This error compensation procedure is efficiently implemented via a weighted least squares formulation where an analytical solution form is obtained. The performance of the developed scheme is extensively tested on 16 benchmark data sets where the results show promising potential of the proposed empirical survival error distribution compensation scheme for binary pattern classification.
Keywords :
least mean squares methods; pattern classification; statistical distributions; binary pattern classification; empirical survival error potential function; error compensation scheme; noise distribution; survival error distribution compensation; weighted least squares scheme; Accuracy; Benchmark testing; Electronic mail; Entropy; Error compensation; Training; Vectors; Binary Classification; Survival Information Potential; Weighted Least Squares;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Control Automation Robotics & Vision (ICARCV), 2014 13th International Conference on
Type :
conf
DOI :
10.1109/ICARCV.2014.7064433
Filename :
7064433
Link To Document :
بازگشت