Domain of attraction on adaptive feature extraction of nonstationary processes

The authors review briefly the procedure relating to the convergence analysis of a learning algorithm for adaptive feature extraction. They then address the issue of identification of a nontrivial domain of attraction for the learning system. The problem is important because such an identification is not only powerful for choosing initial settings of the system, but also holds one of the keys to the quantitative analysis of its adaptivity in a nonstationary environment. The primary results concerning convergence analysis of this algorithm are briefly reviewed.<<ETX>>

[1]  P. Foldiak,et al.  Adaptive network for optimal linear feature extraction , 1989, International 1989 Joint Conference on Neural Networks.

[2]  E. Oja.  Simplified neuron model as a principal component analyzer , 1982, Journal of mathematical biology.

[3]  Ralph Linsker,et al.  Self-organization in a perceptual network , 1988, Computer.

[4]  Terence D. Sanger,et al.  Optimal unsupervised learning in a single-layer linear feedforward neural network , 1989, Neural Networks.

[5]  E. Oja,et al.  On stochastic approximation of the eigenvectors and eigenvalues of the expectation of a random matrix , 1985 .

[6]  Kurt Hornik,et al.  Convergence analysis of local feature extraction algorithms , 1992, Neural Networks.

[7]  Hong Chen,et al.  An on-line unsupervised learning machine for adaptive feature extraction , 1993, 1993 IEEE International Symposium on Circuits and Systems.

[8]  Sun-Yuan Kung,et al.  A neural network learning algorithm for adaptive principal component extraction (APEX) , 1990, International Conference on Acoustics, Speech, and Signal Processing.