Designing compact feedforward neural models with small training data sets

Journal of Aircraft

Volumn 39, Issue 3, 2002, Pages 452-459

  Greenman, Roxana M ^a   Stepniewski, Slawomir W ^b   Jorgensen, Charles C ^c   Roth, Karlin R ^c  

^a LAWRENCE LIVERMORE NATIONAL LABORATORY   (United States)

^b SONY CORPORATION   (Japan)

^c NASA AMES RESEARCH CENTER   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

AERODYNAMICS; BACKPROPAGATION; FEEDFORWARD NEURAL NETWORKS; FUNCTIONS; GENETIC ALGORITHMS; OPTIMIZATION; REGRESSION ANALYSIS;

ACTIVATION FUNCTION; BACKPROPAGATION NEURAL MODELS; BOOTSTRAP TECHNIQUE; REGRESSION MODELS; SMALL TRAINING DATA SETS;

AIRFOILS;

EID: 0036575744     PISSN: 00218669     EISSN: None     Source Type: Journal    
DOI: 10.2514/2.2950     Document Type: Article

References (22)

1. First- and second-order methods for learning: Between steepest descent and Newton's method
2. Two-dimensional high-lift aerodynamic optimization using neural networks
3. High-lift optimization design using neural networks on a multi-element airfoil
4. Minimizing computational data requirements for multi-element airfoils using neural networks
5. Comparison of turbulence models in computing multi-element airfoil flows
6. Manual for the OVERMAGG script system
7. Progress in high-lift aerodynamic calculations
8. User's manual for the HYPGEN hyperbolic grid generator and the HGUI graphical user interface
9. An upwind differencing scheme for the steady state incompressible Navier-Stokes equations
10. Upwind differencing scheme for the time-accurate incompressible Navier-Stokes equations
11. One-equation turbulence model for aerodynamic flows
12. Application of the incompressible Navier-Stokes equations to high-lift flows
13. Method of prediction of wing maximum lift
14. Pruning with generalization based weight saliencies: γOBD, γOBS, advances in neural information processing systems
15. Engineering applications of heuristic multilevel optimization methods
16. Toward a more robust pruning procedure for MLP networks
17. An information-maximization approach to blind separation and blind deconvolution
18. Estimating expected error rates of neural network classifiers in small sample situations: A comparison of cross-validation and bootstrap
19. Bagging predictors