“A Combined Network Architecture Using Art2 and Back Propagation for Adaptive Estimation of Dynamic Processes”
Authors: Einar Sørheim,Affiliation: Institute for Energy Technology (IFE)
Reference: 1990, Vol 11, No 4, pp. 191-199.
Keywords: System identification, nonlinear systems, adaptive control, artificial neural networks, back propagation, ART2
Abstract: A neural network architecture called ART2/BP is proposed. Thc goal has been to construct an artificial neural network that learns incrementally an unknown mapping, and is motivated by the instability found in back propagation (BP) networks: after first learning pattern A and then pattern B, a BP network often has completely ´forgotten´ pattern A. A network using both supervised and unsupervised training is proposed, consisting of a combination of ART2 and BP. ART2 is used to build and focus a supervised backpropagation network consisting of many small subnetworks each specialized on a particular domain of the input space. The ART2/BP network has the advantage of being able to dynamically expand itself in response to input patterns containing new information. Simulation results show that the ART2/BP network outperforms a classical maximum likelihood method for the estimation of a discrete dynamic and nonlinear transfer function.
PDF (847 Kb)
DOI: 10.4173/mic.1990.4.2DOI forward links to this article:
| [1] S. C. Mcloone, S. Mcginnity and G. W. Irwin (2002), doi:10.1080/0020772021000046243 |
| [2] Qiuming Zhu and Ahmed Y. Tawfik (1995), doi:10.1007/BF02312378 |
| [3] Tor A. Johansen and Bjarne A. Foss (1992), doi:10.4173/mic.1992.1.3 |
| [4] Qiuming Zhu and Ahmed Y. Tawfik (1995), doi:10.1016/0952-1976(95)00034-X |
| [5] Tor A. Johansen and Bjarne A. Foss (1995), doi:10.1016/0005-1098(94)00096-2 |
| [6] T.A. Johansen and B.A. Foss (1992), doi:10.1109/SICICI.1992.637617 |
| [7] Tor A. Johansen and Bjarne A. Foss (1994), doi:10.1016/S1474-6670(17)47704-4 |
| [8] Tor A. Johansen and Bjarne A. Foss (1995), doi:10.1007/978-1-4471-3066-6_6 |
[1] CARPENTER, G.A. GROSSBERG, S. (1987). ART2: Self-organization of stable category recognition codes for analog input patterns, Applied Optics, 26, 4919-4930 doi:10.1364/AO.26.004919
[2] CARPENTER, G.A. GROSSBERG, S. (1988). The ART of adaptive pattern recognition by a self-organizing neural network, Computer, 21, 77-88 doi:10.1109/2.33
[3] LJUNG, L. SÖDERSTRÖM (1983). Theory and Practice of Recursive Identification, The MIT press, Cambridge, MA.
[4] RUMELHART, D.E., HINTON, G.E., WILLIAMS, R.J. (1986). Parallel Distributed Processing: Explorations in the Microstructure of Cognition, Vol. 1.The MIT Press, Cambridge, MA.
[5] RYAN, T.W. (1988). The resonance correlation network, Proceedings IJNN88.
BibTeX:
@article{MIC-1990-4-2,
title={{A Combined Network Architecture Using Art2 and Back Propagation for Adaptive Estimation of Dynamic Processes}},
author={Sørheim, Einar},
journal={Modeling, Identification and Control},
volume={11},
number={4},
pages={191--199},
year={1990},
doi={10.4173/mic.1990.4.2},
publisher={Norwegian Society of Automatic Control}
};