On learning the derivatives of an unknown mapping with multilayer feedforward networks

Neural Networks

Volumn 5, Issue 1, 1992, Pages 129-138

  Gallant, A Ronald ^a   White, Halbert ^b  

^a NORTH CAROLINA STATE UNIVERSITY   (United States)

^b UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

Chastic dynamics; Denseness; Estimating derivatives; Sobolev spaces

Indexed keywords

MAPPING; MATHEMATICAL TECHNIQUES - DIFFERENTIATION; SYSTEMS SCIENCE AND CYBERNETICS - HEURISTIC PROGRAMMING;

MULTILAYER FEEDFORWARD NETWORKS; PARTIAL DERIVATIVES; SOBOLEV SPACES;

NEURAL NETWORKS;

EID: 0026449851     PISSN: 08936080     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0893-6080(05)80011-5     Document Type: Article

References (25)

1. Construction of Neural Nets Using the Radon Transform
2. Nonlinear prediction of chaotic time series
3. Approximation by superpositions of a sigmoidal function
4. Lyapunov exponents from time series
5. An elasticity can be estimated consistently without a priori knowledge of functional form
6. On the approximate realization of continuous mappings by neural networks
7. Identification and consistency in seminonparametric regression
8. Identification et Convergence en Regression Semi-Nonparametrique
9. On the asymptotic normality of series estimators when the number of increases and the minimum eigenvalue of X'X/n decreases
10. On fitting a recalcitrant series: The pound/dollar exchange rate, 1974–83
11. There exists a neural network that does not make avoidable mistakes
12. Theory of the back-propagation neural network
13. Multilayer feedforward networks are universal approximators
14. Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks
15. Generic constraints on underspecified target trajectories
16. A numerical approach to ergodic problem of dissipative dynamical systems
17. Universal approximation using feedforward networks with non-sigmoid hidden layer activation functions
18. Cross-validitory choice and assessment of statistical predictions
19. Connectionist nonparametric regression: Multilayer feedforward networks can learn arbitrary mappings