Title :
A Modified Oja–Xu MCA Learning Algorithm and Its Convergence Analysis
Author :
Peng, Dezhong ; Yi, Zhang
Author_Institution :
Comput. Intelligence Lab., Univ. of Electron. Sci. & Technol. of China, Chengdu
fDate :
4/1/2007 12:00:00 AM
Abstract :
The original Oja-Xu minor component analysis (MCA) learning algorithm is not convergent. This brief shows that by modifying Oja-Xu MCA learning algorithm with a normalization step the modified one could be convergent subject to some conditions satisfied. The convergence of the modified MCA learning algorithm is studied by analyzing the convergence of an associated deterministic discrete time system. Necessary and sufficient conditions for convergence are obtained. Simulations further confirm the results
Keywords :
convergence; discrete time systems; eigenvalues and eigenfunctions; learning (artificial intelligence); Oja-Xu MCA learning algorithm; convergence analysis; deterministic discrete time system; eigenvalue; eigenvector; minor component analysis; modified MCA learning algorithm; neural networks; Algorithm design and analysis; Convergence; Data mining; Discrete cosine transforms; Neural networks; Neurons; Signal processing algorithms; Stochastic processes; Sufficient conditions; Vectors; Deterministic discrete time (DDT) system; eigenvalue; eigenvector; minor component analysis (MCA); neural networks;
Journal_Title :
Circuits and Systems II: Express Briefs, IEEE Transactions on
DOI :
10.1109/TCSII.2006.889709
