Fast training of neural trees by adaptive splitting based on cubature

Authors:
B. Llanas;F. J. Sáinz
Affiliations:
Departamento de Matemática Aplicada, ETSI de Caminos, Universidad Politécnica de Madrid, C/Profesor Aranguren s/n, Ciudad Universitaria, 28040 Madrid, Spain;Departamento de Matemática Aplicada, ETSI de Caminos, Universidad Politécnica de Madrid, C/Profesor Aranguren s/n, Ciudad Universitaria, 28040 Madrid, Spain
Venue:
Neurocomputing
Year:
2008

Citing 19
Cited 2

Error estimation in automatic quadrature routines

ACM Transactions on Mathematical Software (TOMS)
Original Contribution: Approximation of a function and its derivative with a neural network

Neural Networks
Original Contribution: On the derivatives of the sigmoid

Neural Networks
Hierarchical mixtures of experts and the EM algorithm

Neural Computation
On the convergence of interior-reflective Newton methods for nonlinear minimization subject to bounds

Mathematical Programming: Series A and B
Approximation of functions by a multilayer perceptron: a new approach

Neural Networks
Hybrid interior point training of modular neural networks

Neural Networks
Combining Artificial Neural Nets: Ensemble and Modular Multi-Net Systems

Combining Artificial Neural Nets: Ensemble and Modular Multi-Net Systems
Noise suppression in training examples for improving generalization capability

Neural Networks
Fast computation in adaptive tree approximation

Numerische Mathematik
Lower Bounds for Simplicial Covers and Triangulations of Cubes

Discrete & Computational Geometry
Efficient Computation of the Hausdorff Distance Between Polytopes by Exterior Random Covering

Computational Optimization and Applications
Constructive approximate interpolation by neural networks

Journal of Computational and Applied Mathematics
Simultaneous Lp-approximation order for neural networks

Neural Networks
Adaptive mixtures of local experts

Neural Computation
Efficient estimation of neural weights by polynomial approximation

IEEE Transactions on Information Theory
Three-dimensional object representation and invariant recognition using continuous distance transform neural networks

IEEE Transactions on Neural Networks
Learning capability and storage capacity of two-hidden-layer feedforward networks

IEEE Transactions on Neural Networks
Real-time learning capability of neural networks

IEEE Transactions on Neural Networks

Edge Detection by Adaptive Splitting

Journal of Scientific Computing
Edge Detection by Adaptive Splitting II. The Three-Dimensional Case

Journal of Scientific Computing

Quantified Score

Hi-index	0.01

Visualization

Abstract

In this paper we prove that any affine function defined on a d-simplex in R^d can be uniformly approximated by a single-layer neural network having only two neurons irrespective of d. The weights of this network are obtained in a closed analytical form, without training. This fact gives a correspondence rule that allows to transform mathematical approximants based on piecewise affine functions, into neural networks. We introduce such an approximant, adaptive splitting based on cubature (ASBC), for the efficient approximation of continuous functions. Using ASBC and the above correspondence rule, we obtain a neural tree. Numerical experiments on learning the function distance from a variable point to a geometric body in two and three dimensions show fast learning speed and high accuracy when compared with single-hidden layer feedforward networks trained by a trust region method based on the interior-reflective Newton algorithm.