A New Approach Using Temporal Radial Basis Function in Chronological Series

In this paper, we present an extended form of Radial Basis Function network called Temporal-RBF (T-RBF) network. This extended network can be used in decision rules and classification in Spatio-Temporal domain applications, like speech recognition, economic fluctuations, seismic measurements and robotics applications. We found that such a network complies with relative ease to constraints such as capacity of universal approximation, sensibility of node, local generalisation in receptive field, etc. For an optimal solution based on a probabilistic approach with a minimum of complexity, we propose two TRBF models (1 and 2). Application to the problem of Mackey-Glass time series has revealed that TRBF models are very promising, compared to traditional networks.

Anahtar Kelimeler:

Temporal RBF, Classification, Spatio-Temporal, Speech recognition, Robotics applications

A New Approach Using Temporal Radial Basis Function in Chronological Series

Keywords:

Temporal RBF, Classification, Spatio-Temporal, Speech recognition, Robotics applications,

PDF

___

S.P. Day, M.R. Davenport, Continuous-Time temporal Back-Propagation with adaptable Time Delays. IEEE Transaction on Neural Networks 1993; 4(12):348–354.
D.T. Lin, The adaptive Time delay Neural Network Characterization and Application to Pattern Recognition, Prediction and signal processing. Thesis Report PHD. University of Maryland, 1994.
C. Wohler, J.K. Anlauf, Real time object recognition on image sequences with adaptable time delay Neural Network Algorithm -application to autonomous vehicles. Image and Vision, Computing Journal 2001; 19(9–10): 593–618.
R. Bonne, M. Cruciano, J.P.A. De Beauville, An algorithm for the addition of time-delayed connections to recurrent neural network. ESANN’2000 proceeding, Bruge (Belgium); 293–298.
F. Gers, D. Eck, J. Schmidhuber, Applying LSTM to time series predictable through time window approaches. Technical Report; IDSIA-IDSIA-22-00. Institute Dalle Molle di studi Sull intelligenza artiŞciale, Galleria2, Switzerland, 1999.
N.B. Karyianis, Reformulated radial basis function neural network trained by gradient descent. IEEE transaction on Neural Network may 1999; 10(3): 657–669.
S. Haykin, Neural Networks a comprehensive foundation. Prentice Hall Upper Saddle River, New Jersey, 1999. [8] G. Zheng, S. Billings, Radial basis function network conŞguration using mutual information and the Orthogonal Least Squares algorithms. Neural Network 1996; 9(6): 1619–1637.
Z. Mekkakia, L. Mesbahi, M. Lakehal, Initialising of RBF centers by SOM Algorithm. MS’2002, International Conference on modeling and simulation in technical and social sciences, Girona, Catalonia, Spain 25–27 june 2002; 717–723.
D.T. Lin, J.E. Dayhoﬀ, Network Unfolding Algorithm and universal spatio-temporal function approximation. Technical Research Report TR95-6. Institute for system research ISR; University of Maryland; College Park MD.
M.K. Titsias, A.C. Likas, Shared kernel model for class Conditional density Estimation. IEEE transaction on Neural Network September 2001; 12(5): 987–996.