Abstract
The OWE (Orthogonal Weight Estimator) architecture is constituted of a main MLP in which the values of each weight is computed by another MLP (an OWE). The number of OWEs is equal to the number of weights of the main MLP. But the computation of each OWE is done independently. Therefore the training and relaxation phases can straightforward parallelized. We report the implementation of this architecture on an Intel Paragon parallel computer and the comparison with its implementation on a sequential computer.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Anguita, D., Parodi G. and Zunino R.: An Efficient inplementation of BP on RISC-based workstations. Neurocomputing 6 (1994) 57–65, Elsevier.
Arbib, M.A., Bekey, G., Lyons, D., Sun, R.: Neural architectures and distributed AI. Workshp proceedings, Center for Neural Engineering, University of Southern, California, October 19–20, 1993.
Barbosa, Valmir C: Massively parallel models of computation: distributed parallel processing in artificial intelligence and optimisation. Ellis horwood series in Artificial intelligence (col.), Ellis Horwood (ed.), New York, London, Toronto. XII-253 (1993)
Kerchoffs, E.J.H, Wedman, F.W. and Frietman, E.E.E.: Speeding up backpropagation training on hypercube computer. Neurocomputing 4 (1992) 43–63
Pican, N., Fort, J.C, and Alexandre, F.: Lateral Contribution Learning Algorithm for Multi-MLP Architecture. ESANN'94 Proceedings, D Facto Brussels Belgium. April 20–22, 1994.
Pican, N., Fort, J.C, and Alexandre, F.: An On-line Learning Algorithm for the Orthogonal Weight Estimation of MLP. Neural Processing Letters, D Facto Brussels Belgium. 1(1) (1994), 21–24.
Pican, N.: Synaptic efficiency modulations for context integration: The Meta ODWE architecture. In ESANN'96 Proceedings, D Facto Brussels Belgium. Bruges April 24–26, 1996.
Pican, N.: An Orthogonal Delta Weight Estimator for MLP Architectures. In ICNN'96 Proceedings, Washington DC. June 2–6, 1996.
Pican, N., Fohr, D., and Mari, J.F HMMs and OWE Neural Network for Continuous Speech Recognition. In Proceedings of International Conference on Spoken Language Processing, ICSLP. Philadelphia October 3–6, 1996.
Singer, A.: Implementations of artificial neural networks on the connection machine. Parallel Computing 14 (1990) 305–315
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Pican, N. (1996). Intrinsic and parallel performances of the OWE neural network architecture. In: von der Malsburg, C., von Seelen, W., Vorbrüggen, J.C., Sendhoff, B. (eds) Artificial Neural Networks — ICANN 96. ICANN 1996. Lecture Notes in Computer Science, vol 1112. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-61510-5_127
Download citation
DOI: https://doi.org/10.1007/3-540-61510-5_127
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-61510-1
Online ISBN: 978-3-540-68684-2
eBook Packages: Springer Book Archive