Initialization by selection for wavelet network training

Neurocomputing

Volumn 34, Issue 1-4, 2000, Pages 131-143

  Oussar, Yacine ^a   Dreyfus, Gérard ^a  

^a PSL RESEARCH UNIVERSITY   (France)

Author keywords

Initializing parameters; Nonlinear static modeling; Training; Wavelet networks

Indexed keywords

COMPUTER SIMULATION; HEURISTIC METHODS; LEARNING ALGORITHMS; LEARNING SYSTEMS; PARAMETER ESTIMATION; WAVELET TRANSFORMS;

INITIALIZING PARAMETER; NETWORK TRAINING; NONLINEAR STATIC MODELING; WAVELET NETWORKS;

FEEDFORWARD NEURAL NETWORKS;

ARTICLE; ARTIFICIAL NEURAL NETWORK; CONTROLLED STUDY; HISTOGRAM; MATHEMATICAL COMPUTING; METHODOLOGY; MODEL; PARAMETER; PRIORITY JOURNAL; TRAINING; WAVEFORM;

EID: 0034284052     PISSN: 09252312     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0925-2312(00)00295-2     Document Type: Article

References (13)

1. Solving linear least squares problems by gram-schmidt orthogonalization
2. Orthogonal least squares methods and their application to non-linear system identification
3. Regression modeling in back-propogation and projection pursuit learning
4. Wavelets in identification: Wavelets, splines, neurons, fuzzies: How good for identification?
5. Multidimensional wavelet frames
6. A theory for multiresolution signal decomposition: The wavelet transform
7. Training wavelet networks for nonlinear dynamic input-output modeling
8. Analysis and synthesis of feedforward neural networks using discrete affine wavelet transformations
9. Wavelet networks
10. Using wavelet network in nonparametric estimation