Neural Networks, Brazilian Symposium on (2010)

Sao Bernardo do Campo, Sao Paulo Brazil

Oct. 23, 2010 to Oct. 28, 2010

ISBN: 978-0-7695-4210-2

pp: 25-30

DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/SBRN.2010.13

ABSTRACT

This paper presents a new probabilistic neural network model, called IPNN (for Incremental Probabilistic Neural Network), which is able to learn continuously probability distributions from data flows. The proposed model is inspired in the Specht's general regression neural network, but have several improvements which makes it more suitable to be used in on-line and robotic tasks. Moreover, IPNN is able to automatically define the network structure in an incremental and on-line way, with new units added whenever necessary to represent new training data. The experiments performed using the proposed model shows that IPNN is able to approximate continuous functions using few probabilistic units.

INDEX TERMS

Probabilistic neural networks, General regression neural networks, Incremental learning, Gaussian mixture models, Semi-parametric methods, Bayesian methods

CITATION

M. R. Heinen and P. M. Engel, "IPNN: An Incremental Probabilistic Neural Network for Function Approximation and Regression Tasks,"

*2010 Eleventh Brazilian Symposium on Neural Networks (SBRN 2010)(SBRN)*, Sao Paulo, 2010, pp. 25-30.

doi:10.1109/SBRN.2010.13

CITATIONS