Optimizing number of hidden neurons in neural networks

Authors:
Yinyin Liu;Janusz A. Starzyk;Zhen Zhu
Affiliations:
School of Electrical Engineering and Computer Science, Ohio University, Athens, OH;School of Electrical Engineering and Computer Science, Ohio University, Athens, OH;School of Electrical Engineering and Computer Science, Ohio University, Athens, OH
Venue:
AIAP'07 Proceedings of the 25th conference on Proceedings of the 25th IASTED International Multi-Conference: artificial intelligence and applications
Year:
2007

Citing 7
Cited 3

Multilayer feedforward networks are universal approximators

Neural Networks
Introduction to artificial neural systems

Introduction to artificial neural systems
Neural networks and the bias/variance dilemma

Neural Computation
Early Stopping-But When?

Neural Networks: Tricks of the Trade, this book is an outgrowth of a 1996 NIPS workshop
Feedforward Neural Network Construction Using Cross Validation

Neural Computation
Asymptotic statistical theory of overtraining and cross-validation

IEEE Transactions on Neural Networks
Linear-least-squares initialization of multilayer perceptrons through backpropagation of the desired response

IEEE Transactions on Neural Networks

Neural network architecture selection: can function complexity help?

Neural Processing Letters
Creating and measuring diversity in multiple classifier systems using support vector data description

Applied Soft Computing
Extension of the generalization complexity measure to real valued input data sets

ISNN'10 Proceedings of the 7th international conference on Advances in Neural Networks - Volume Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, a novel and effective criterion based on the estimation of the signal-to-noise-ratio figure (SNRF) is proposed to optimize the number of hidden neurons in neural networks to avoid overfitting in the function approximation. SNRF can quantitatively measure the useful information left unlearned so that overfitting can be automatically detected from the training error only without use of a separate validation set. It is illustrated by optimizing the number of hidden neurons in a multi-layer perceptron (MLP) using benchmark datasets. The criterion can be further utilized in the optimization of other parameters of neural networks when overfitting needs to be considered.