Pi-DTB, Discrete Time Backpropagation with Product Units

This paper introduces a neural network that combines the power of two different approaches to obtaining more efficient neural structures for processing complex real signals: the use of trainable temporal delays in the synapses and the inclusion of product terms within the combination function. In addition to the neural network structure itself, the paper presents a new algorithm for training this particular type of networks and provides a set of examples using chaotic series, which compare the results obtained by these networks and training algorithm to other structures.

[1]  Richard J. Duro,et al.  Discrete-time backpropagation for training synaptic delay-based artificial neural networks , 1999, IEEE Trans. Neural Networks.

[2]  José Mira Mira,et al.  NETTOOL: A Hybrid Connectionist-Symbolic Development Environment , 1995, IWANN.

[3]  Francisco Sandoval,et al.  From Natural to Artificial Neural Computation , 1995 .

[4]  Franz Pichler,et al.  Computer Aided Systems Theory — EUROCAST '91 , 1991, Lecture Notes in Computer Science.

[5]  F. Takens.  Detecting strange attractors in turbulence , 1981 .

[6]  R. Mañé,et al.  On the dimension of the compact invariant sets of certain non-linear maps , 1981 .

[7]  Richard J. Duro,et al.  Training Higher Order Gaussian Synapses , 1999, IWANN.

[8]  Colin Giles,et al.  Learning, invariance, and generalization in high-order neural networks. , 1987, Applied optics.

[9]  Donald E. Waagen,et al.  Evolving recurrent perceptrons for time-series modeling , 1994, IEEE Trans. Neural Networks.

[10]  José Mira Mira,et al.  Linear and Algorithmic Formulation of Co-operative Computation in Neural Nets , 1991, EUROCAST.

[11]  Tommy W. S. Chow,et al.  Performance enhancement using nonlinear preprocessing , 1996, IEEE Trans. Neural Networks.

[12]  José Mira,et al.  Foundations and Tools for Neural Modeling , 1999, Lecture Notes in Computer Science.

[13]  Marios M. Polycarpou,et al.  High-order neural network structures for identification of dynamical systems , 1995, IEEE Trans. Neural Networks.

[14]  Floris Takens,et al.  On the numerical determination of the dimension of an attractor , 1985 .