Nonlinear system modeling via knot-optimizing B-spline networks

IEEE Transactions on Neural Networks

Volumn 12, Issue 5, 2001, Pages 1013-1022

  Yiu, Ka Fai Cedric ^b   Wang, Song ^c   Teo, Kok Lay ^a,b   Tsoi, Ah Chung ^d  

^a IEEE

^b HONG KONG POLYTECHNIC UNIVERSITY   (Hong Kong)

^c UNIVERSITY OF WESTERN AUSTRALIA   (Australia)

^d UNIVERSITY OF WOLLONGONG   (Australia)

Author keywords

B splines; Knot points; Neural network; Nonlinear system modeling

Indexed keywords

APPROXIMATION THEORY; LEARNING ALGORITHMS; LEARNING SYSTEMS; MATHEMATICAL MODELS; NONLINEAR SYSTEMS; PERTURBATION TECHNIQUES; POLYNOMIALS; RANDOM PROCESSES; SIMULATED ANNEALING; TENSORS;

B-SPLINE NETWORKS; CUBIC SPLINE FUNCTION; KNOT POINTS; NONLINEAR SYSTEM MODELING;

RADIAL BASIS FUNCTION NETWORKS;

EID: 0035441376     PISSN: 10459227     EISSN: None     Source Type: Journal    
DOI: 10.1109/72.950131     Document Type: Article

References (18)

1. A new look at the statistical model identification
2. Thermodynamical approach to the traveling salesman problem: An efficient simulation algorithm
3. Multivariate adaptive regression splines
4. Placing spline knots in neural networks using splines as activation functions
5. ASMOD - An algorithm for adaptive spline modeling of observation data
6. Optimization by simulated annealing
7. Multilayer feedforward networks are universal approximators
8. Learning with localized receptive fields
9. Modeling by shortest data description
10. Approximation of nonlinear systems with radial basis function neural networks
11. Universal approximation using feedforward neural networks: A survey of some existing methods, and some new results
12. Discrete-time recurrent neural network architectures: A unifying review
13. Recurrent neural networks: A constructive algorithm, and its properties