Improving the MLP learning by using a method to calculate the initial weights of the network based on the quality of similarity measure

Authors:
Yaima Filiberto Cabrera;Rafael Bello Pérez;Yailé Caballero Mota;Gonzalo Ramos Jimenez
Affiliations:
Department of Computer Sciences, University of Camaguey, Cuba;Department of Computer Sciences, Central University of Las Villas, Cuba;Department of Computer Sciences, University of Camaguey, Cuba;Department of Languages and Computer Science, University of Malaga, Spain
Venue:
MICAI'11 Proceedings of the 10th international conference on Artificial Intelligence: advances in Soft Computing - Volume Part II
Year:
2011

Citing 15
Cited 1

Multilayer feedforward networks are universal approximators

Neural Networks
Initializing back propagation networks with prototypes

Neural Networks
Estimating attributes: analysis and extensions of RELIEF

ECML-94 Proceedings of the European conference on machine learning on Machine Learning
A New Weight Initialization Method for the MLP with the BP inMulticlass Classification Problems

Neural Processing Letters
Machine Learning

Machine Learning
A Generalized Definition of Rough Approximations Based on Similarity

IEEE Transactions on Knowledge and Data Engineering
A study of the behavior of several methods for balancing machine learning training data

ACM SIGKDD Explorations Newsletter - Special issue on learning from imbalanced datasets
Statistical Comparisons of Classifiers over Multiple Data Sets

The Journal of Machine Learning Research
Accelerated Gradient Learning Algorithm for Neural Network Weights Update

KES '08 Proceedings of the 12th international conference on Knowledge-Based Intelligent Information and Engineering Systems, Part I
An Evolutionary Approach for Tuning Artificial Neural Network Parameters

HAIS '08 Proceedings of the 3rd international workshop on Hybrid Artificial Intelligence Systems
Revisiting the Problem of Weight Initialization for Multi-Layer Perceptrons Trained with Back Propagation

Advances in Neuro-Information Processing
Ideas about a regularized MLP classifier by means of weight decay stepping

ICANNGA'09 Proceedings of the 9th international conference on Adaptive and natural computing algorithms
A resource limited immune approach for evolving architecture and weights of multilayer neural network

ICSI'10 Proceedings of the First international conference on Advances in Swarm Intelligence - Volume Part I
Simultaneous evolution of neural network topologies and weights for classification and regression

IWANN'05 Proceedings of the 8th international conference on Artificial Neural Networks: computational Intelligence and Bioinspired Systems
High-order and multilayer perceptron initialization

IEEE Transactions on Neural Networks

A method for building prototypes in the nearest prototype approach based on similarity relations for problems of function approximation

MICAI'12 Proceedings of the 11th Mexican international conference on Advances in Artificial Intelligence - Volume Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

This work presents a technique that integrates the backpropagation learning method with a method to calculate the initial weights in order to train the Multilayer Perceptron Model. The method to calculate the initial weights of the MLP is based on the quality of similarity measure proposed on the framework of the extended Rough Set Theory. Experimental results show that the proposed initialization method performs better than other methods used to calculate the weight of the features, so it is an interesting alternative to the conventional random initialization.