Abstract
A new constructive algorithm is presented for building neural networks that learn to reproduce output temporal sequences based on one or several input sequences. This algorithm builds a network for the task of system modelling, dealing with continuous variables in the discrete time domain. The constructive scheme makes it user independent. The network's structure consists of an ordinary set and a classification set, so it is a hybrid network like that of Stokbro et al. [6], but with a binary classification. The networks can easily be interpreted, so the learned representation can be transferred to a human engineer, unlike many other network models. This allows for a better understanding of the system structure than just its simulation. This constructive algorithm limits the network complexity automatically, hence preserving extrapolation capabilities. Examples with real data from three totally different sources show good performance and allow for a promising line of research.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Fahlman SE. The Recurrent Cascade-Correlation architecture. CMU-CS-91-100, Carnegie Mellon University, 1991
Fahlman SE, Lebiere C. The Cascade-Correlation learning architecture. Advancces in Neural Information Processing Systems II (Denver, 1989), ed. DS Touretzky, Morgan-Kaufmann, San Mateo, CA, 1990, 524–532
Moody J, Darken C. Fast learning in networks of locally-tuned processing units. Neural Computation 1989; 1: 281–294
Back AD, Tsoi AC. FIR and IIR synapses, a new neural network architecture for time series modeling. Neural Computation 1991; 3: 375–385
Jordan MI. Attractor dynamics and parallelism in a connectionist sequential machine. Proc. 8th Ann Conf Cognitive Science Study. Erlbaum, Hillsdale, NJ, 1986, 531–546
Stokbro K, Umberger DK, Hertz JA. Exploiting neurons with localized receptive fields to learn chaos. Preprint 90/28 S, Nordita, Copenhagen, Denmark, 1990
Jones RD, Lee YC, Barnes CW, Flake GW, Lee K, Lewis PS, Qian S. Function approximation and time series prediction with neural networks. Los Alamos National Laboratory Report LA-UR-90-21, 1990
Moody J, Darken C. Learning with localized receptive fields. Pro Connectionist Models Summer School (Pittsburgh, 1988), eds D Touretzky, G Hinton, T Sejnowski, Morgan-Kaufmann, San Mateo, CA, 1988, 133–143
Fahlman SE. An empirical study of learning speed in back-propagation networks. CMU-CS-88-162, 1988
Crespo JL. Procesamiento paralelo y distribuido aplicado a la simulación de sistemas. PhD Thesis, University of Cantabria, 1992
Burnash RJC, Ferral RL, McGuire RA. A generalized streamflow simulation system. Conceptual Modeling for Digital Computers. 1979
Box GEP, Jenkins GM. Time series analysis: forecasting and control (Rev. ed.) Holden Day, San Francisco, CA, 1976
Liu L-M. Box-Jenkins time series analysis. BMDP Statistical Software Manual, Vol. 1. University of California Press, 1990, 435–488
Crespo JL, Mora E, Peire J. Estimation performance of neural networks. ISIE'93, Budapest, Hungary, 1993
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Crespo, J.L., Mora, E. A PDP constructive algorithm for system modelling. Neural Comput & Applic 4, 175–182 (1996). https://doi.org/10.1007/BF01414878
Issue Date:
DOI: https://doi.org/10.1007/BF01414878