DocumentCode
2104508
Title
Strong and weak stability of randomized learning algorithms
Author
Ke Luo ; Zhiyang Jia ; Wei Gao
Author_Institution
Shaoyang Univ. Libr., Shaoyang, China
fYear
2012
fDate
9-11 Nov. 2012
Firstpage
887
Lastpage
891
Abstract
An algorithm is called stable at a training set S if any change of a single point in S yields only a small change in the output. Stability of the learning algorithm is necessary for learnability in the supervised classification and regression setting. In this paper, we give formal definitions of strong and weak stability for randomized algorithms and prove non-asymptotic bounds on the difference between the empirical and expected error.
Keywords
learning (artificial intelligence); randomised algorithms; regression analysis; formal definitions; learnability; nonasymptotic bounds; randomized algorithms; randomized learning algorithms; regression setting; strong stability; supervised classification; training set; weak stability; empirical error; generalization error; randomized learning algorithms; strong stability; weak stability;
fLanguage
English
Publisher
ieee
Conference_Titel
Communication Technology (ICCT), 2012 IEEE 14th International Conference on
Conference_Location
Chengdu
Print_ISBN
978-1-4673-2100-6
Type
conf
DOI
10.1109/ICCT.2012.6511323
Filename
6511323
Link To Document