Corpus ID: 15691523

NEURAL NETWORKS FOR OPTIMAL APPROXIMATION OF SMOOTH

@inproceedings{Mhaskar1996NEURALNF,
  title={NEURAL NETWORKS FOR OPTIMAL APPROXIMATION OF SMOOTH},
  author={H. Mhaskar},
  year={1996}
}
We prove that neural networks with a single hidden layer are capable of providing an optimal order of approximation for functions assumed to possess a given number of derivatives, if the activation function evaluated by each principal element satisfies certain technical conditions. Under these conditions, it is also possible to construct networks that provide a geometric order of approximation for analytic target functions. The permissible activation functions include the squashing function (1… Expand
7 Citations
Time-delay neural networks, volterra series, and rates of approximation
  • I. Sandberg
  • ISCAS '98. Proceedings of the 1998 IEEE International Symposium on Circuits and Systems (Cat. No.98CH36187)
  • 1998
Generalization bounds for function approximation from scattered noisy data
  • 82
Neural Networks with Local Receptive Fields and Superlinear VC Dimension
  • M. Schmitt
  • Mathematics, Computer Science
  • Neural Computation
  • 2002
  • 9
Algebraic Analysis for Non-identifiable Learning Machines
  • 48
  • Highly Influenced

References

SHOWING 1-10 OF 25 REFERENCES
Neural networks for localized approximation of real functions
  • H. Mhaskar
  • Mathematics
  • Neural Networks for Signal Processing III - Proceedings of the 1993 IEEE-SP Workshop
  • 1993
  • 31
Multilayer Feedforward Networks with a Non-Polynomial Activation Function Can Approximate Any Function
  • 1,327
  • Highly Influential
  • PDF
Universal Approximation Using Radial-Basis-Function Networks
  • 3,292
  • PDF
Approximation properties of a multilayered feedforward artificial neural network
  • H. Mhaskar
  • Mathematics, Computer Science
  • Adv. Comput. Math.
  • 1993
  • 130
Approximation by superpositions of a sigmoidal function
  • G. Cybenko
  • Mathematics, Computer Science
  • Math. Control. Signals Syst.
  • 1989
  • 4,230
  • PDF
Dimension-independent bounds on the degree of approximation by neural networks
  • 73
Universal approximation bounds for superpositions of a sigmoidal function
  • A. Barron
  • Mathematics, Computer Science
  • IEEE Trans. Inf. Theory
  • 1993
  • 2,360
  • PDF
Regularization Theory and Neural Networks Architectures
  • 1,307
Multilayer feedforward networks are universal approximators
  • 15,473
  • PDF
From regularization to radial, tensor and additive splines
  • 105
  • PDF
...
1
2
3
...