Neural networks with asymmetric activation function for function approximation
GSS Gomes, TB Ludermir… - 2009 International Joint …, 2009 - ieeexplore.ieee.org
The choice of activation functions may strongly influence complexity and performance of
neural networks. However a limited number of activation functions have been used in …
neural networks. However a limited number of activation functions have been used in …
On dimension-independent approximation by neural networks and linear approximators
S Giulini, M Sanguineti - Proceedings of the IEEE-INNS-ENNS …, 2000 - ieeexplore.ieee.org
Sets of multivariable functions that can be approximated with" dimension-independent" rates
either by linear approximators or by neural networks having various types of computational …
either by linear approximators or by neural networks having various types of computational …
On the (1+ 1/2) layer neural networks as universal approximators
I Ciuca, JA Ware - 1998 IEEE International Joint Conference on …, 1998 - ieeexplore.ieee.org
Deals with the approximation of continuous functions by feedforward neural networks. After
presenting one of the main results of Ito, the paper tries to get a universal approximator …
presenting one of the main results of Ito, the paper tries to get a universal approximator …
How to choose an activation function
HN Mhaskar, CA Micchelli - Advances in neural information …, 1993 - proceedings.neurips.cc
We study the complexity problem in artificial feedforward neural networks designed to
approximate real valued functions of several real variables; ie, we estimate the number of …
approximate real valued functions of several real variables; ie, we estimate the number of …
Asymptotic approximation power for neural networks
PJSG Ferreira, AJ Pinho - 1998 IEEE International Joint …, 1998 - ieeexplore.ieee.org
Asymptotic approximation power for neural networks Page 1 Asymptotic approximation power
for neural networks Paul0 J. S. G. Ferreira, Armando J. Pinho Abstract This paper studies the …
for neural networks Paul0 J. S. G. Ferreira, Armando J. Pinho Abstract This paper studies the …
The probability characteristic of function approximation based on artificial neural network
H Gao - … International Conference on Computer Application and …, 2010 - ieeexplore.ieee.org
In this paper, a desirable BP neural network is established for function approximation, and
three kinds of concrete function is approximated by the network. The better cases for function …
three kinds of concrete function is approximated by the network. The better cases for function …
Automatic sizing of neural networks for function approximation
Neural networks (NN) are a very efficient and powerful function approximation tool. Inspired
by the brain structure and functions, NN are usually trained with backpropagation learning …
by the brain structure and functions, NN are usually trained with backpropagation learning …
Comparison of rates of linear and neural network approximation
V Kurková, M Sanguineti - Proceedings of the IEEE-INNS …, 2000 - ieeexplore.ieee.org
We develop some mathematical tools for comparison of rates of fixed versus variable basis
function approximation. Using these tools, we describe sets of multivariable functions, for …
function approximation. Using these tools, we describe sets of multivariable functions, for …
Function approximation using a partition of the input space
P Koiran - … 1992] IJCNN International Joint Conference on …, 1992 - ieeexplore.ieee.org
Feedforward neural networks can uniformly approximate continuous functions. It is shown
that a simple geometric proof of this theorem, proposed originally for networks of Heaviside …
that a simple geometric proof of this theorem, proposed originally for networks of Heaviside …
The general approximation theorem
A general approximation theorem is proved. It uniformly envelopes both the classical Stone
theorem and approximation of functions of several variables by means of superpositions and …
theorem and approximation of functions of several variables by means of superpositions and …