Title :
A class of PCA learning algorithms and their convergence
Author_Institution :
Dept. of Comput. Eng., Chengdu Aeronaut. Vocational & Tech. Coll., Chengdu, China
Abstract :
This paper proposes a class of principal component analysis (PCA) learning algorithms with constant learning rates. It will prove via deterministic discrete time (DDT) method that these PCA learning algorithms are globally convergent.
Keywords :
convergence; learning (artificial intelligence); neural nets; principal component analysis; PCA learning algorithms; constant learning rates; convergence; deterministic discrete time method; principal component analysis;
Conference_Titel :
Advanced Computer Theory and Engineering (ICACTE), 2010 3rd International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-6539-2
DOI :
10.1109/ICACTE.2010.5579030
