Title :
Using recursive least square learning method for principal and minor components analysis
Author :
Wong, A.S.Y. ; Wong, K.W. ; Leung, C.S.
Author_Institution :
Dept. of Electr. Electron., Hong Kong Univ., Hong Kong
Abstract :
In combining principal and minor components analysis, a parallel extraction method based on the recursive least square algorithm is suggested to extract the principal components of the input vectors. After the extraction, the error covariance matrix obtained in the learning process is used to perform minor components analysis. The minor components found are then pruned so as to achieve a higher compression ratio. Simulation results show that both the convergent speed and the compression ratio are improved, which in turn indicate that our method effectively combines the extraction of the principal components and the pruning of the minor components
Keywords :
convergence of numerical methods; covariance matrices; data compression; feature extraction; image reconstruction; learning (artificial intelligence); least squares approximations; parallel processing; recursive estimation; compression ratio; convergent speed; error covariance matrix; feedforward neural networks; image reconstruction; input vectors; learning process; minor components analysis; minor components pruning; parallel extraction method; principal components analysis; recursive least square algorithm; recursive least square learning method; simulation results; Computational complexity; Convergence; Covariance matrix; Equations; Learning systems; Least squares methods; Neural networks; Neurons; Principal component analysis; Resonance light scattering;
Conference_Titel :
Acoustics, Speech and Signal Processing, 1998. Proceedings of the 1998 IEEE International Conference on
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-4428-6
DOI :
10.1109/ICASSP.1998.675458
