Improving RBF-DDA Performance on Optical Character Recognition through Parameter Selection

Authors:
A. L. I. Oliveira;F. B. L. Neto;S. R. L. Meira
Affiliations:
Pernambuco University, Brazil;Pernambuco University, Brazil;Federal University of Pernambuco, Brazil
Venue:
ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 4 - Volume 04
Year:
2004

Citing 0
Cited 5

Letters: Improved pruning strategy for radial basis function networks with dynamic decay adjustment

Neurocomputing
Letter: Improving constructive training of RBF networks through selective pruning and model selection

Neurocomputing
Spatially adaptive sparse grids for high-dimensional data-driven problems

Journal of Complexity
A comparative study on support vector machine and constructive RBF neural network for prediction of success of dental implants

CIARP'05 Proceedings of the 10th Iberoamerican Congress conference on Progress in Pattern Recognition, Image Analysis and Applications
A comparative study on machine learning techniques for prediction of success of dental implants

MICAI'05 Proceedings of the 4th Mexican international conference on Advances in Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

The Dynamic Decay Adjustment (DDA) algorithm is a fast constructive algorithm for training RBF neural networks. In previous works it has been shown that for some datasets the generalization performance of RBF-DDA depends only weakly on the algorithm parameters 驴+ and 驴-. However, we have observed experimentally that for some problems performance is considerably dependent on the value of 驴-. In this work we propose a method for selecting the value of 驴- for performance optimization. The proposed method has been evaluated on three optical recognition datasets from the UCI repository. The results show that the proposed method considerably improves the performance of RBF-DDA with default parameters on these tasks. The results are compared to MLP and k-NN results obtained in previous works. It is shown that the method proposed in this paper outperforms MLPs and obtains results comparable to k-NN on these tasks.