DocumentCode :
50849
Title :
Parameterized AdaBoost: Introducing a Parameter to Speed Up the Training of Real AdaBoost
Author :
Shuqiong Wu ; Nagahashi, Hiroshi
Author_Institution :
Dept. of Comput. Intell. & Syst. Sci., Tokyo Inst. of Technol., Yokohama, Japan
Volume :
21
Issue :
6
fYear :
2014
fDate :
Jun-14
Firstpage :
687
Lastpage :
691
Abstract :
As a machine learning algorithm, AdaBoost has obtained considerable success in data classification and object detection. Later its generalized version called Real AdaBoost was proposed by Schapire and Singer. Real AdaBoost increases weights for misclassified samples and decreases weights for correctly classified samples in every iteration. This kind of weight adjustment focuses on the samples with large weights and tries to make them correctly classified in future runs. However, it may lead to the misclassification of some other samples that have been correctly classified in previous runs. If we can curb this kind of misclassification during the boosting process, a faster training can be achieved. Based on this assumption, we propose Parameterized AdaBoost in which a parameter is devised to penalize the misclassification of samples that have already been correctly classified. Then we analyse that samples with positive classification margins in Parameterized AdaBoost are more than in Real AdaBoost. Experimental results show that our approach achieves a faster convergence of the training error and also improves the generalization error to some degree when compared with Real AdaBoost.
Keywords :
learning (artificial intelligence); pattern classification; AdaBoost training; boosting process; classification margins; data classification; generalization error; machine learning algorithm; object detection; parameterized AdaBoost; Indexes; Machine learning algorithms; Materials; Object detection; Signal processing algorithms; Training; Training data; AdaBoost; generalization error; machine learning; real AdaBoost; training error;
fLanguage :
English
Journal_Title :
Signal Processing Letters, IEEE
Publisher :
ieee
ISSN :
1070-9908
Type :
jour
DOI :
10.1109/LSP.2014.2313570
Filename :
6778011
Link To Document :
بازگشت