Title :
Learning rate and outlier analysis of linear learning algorithms
Author :
Yin, Hongfeng ; Klasa, Stan
Author_Institution :
Dept. of Comput. Sci., Concordia Univ., Montreal, Que., Canada
Abstract :
The learning rate is analyzed for linear learning algorithms in this paper. In the presence of outliers, the robustness of several linear learning algorithms is given and it is shown that an absolute criterion based learning algorithm is more robust than the corresponding quadratic criterion based learning algorithm
Keywords :
convergence; learning (artificial intelligence); neural nets; pattern recognition; statistical analysis; learning rate; linear learning algorithms; outlier analysis; robustness; Algorithm design and analysis; Approximation algorithms; Computer science; Convergence; Difference equations; Differential equations; Neural networks; Principal component analysis; Robustness; Stochastic processes;
Conference_Titel :
Neural Networks, 1995. Proceedings., IEEE International Conference on
Conference_Location :
Perth, WA
Print_ISBN :
0-7803-2768-3
DOI :
10.1109/ICNN.1995.488183
