Abstract
In this paper we study the parallelization of the inference process for connectionist models. We use a symbolic formalism for the representation of the connectionist models. With this translation, the training mechanism is local in the elements of the network, the computing power is improved in the network nodes and a local hybridization with symbolic parts is achieved. The inference in the final knowledge network can be parallelized, whether the knowledge corresponds to a symbolic module, a connectionist model or a hybrid connectionist-symbolic module. Besides, the concurrency for knowledge networks corresponding to connectionist models is presented for the phases of processing and training. The parallelization is studied for a multiprocessor architecture with shared memory.
This research was supported in part by the Government of Galicia (Spain), grant XUGA10503B/96.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
H. Chung, J-H. Yoon and S.R. Maeng. A sistolic array exploiting the inherent parallelism of artificial neural networks. Microprocessing and Microprogramming. 33(3):145–159, 1992.
M. Heileman, G.L. Georgiopoulos and Roome W.D. A general framework for concurrent simulation of neural network models. IEEE Transactions on Software Engineering, 18(7):551–562, 1992.
T. Lallement, Y. Cornu and Vialle S. An abstract machine for implementing connectionist and hybrid systems on multi-processor architectures. Parallel Processing for Artificial Intelligence, Machine Intelligence and Pattern Recognition Series. H. Kitano, V. Kumer and C.B. Sutter (Eds.), Elsevier Science, (2), 1994.
T.E. Lange. Simulation of heterogeneous neural networks on serial and parallel machines. Parallel Computing, (4):287–303, 1990.
Cabarcos M., Otero M., Cabalar P., and Otero R.P. Efficient concurrent execution of medtool expert systems. EXPERSYS-96: Expert systems applications and artificial intelligence. Technology Transfer series. A. Niku-Lari (Ed) pages 389–394 1996.
The medtool project and related activities are described in web documents http://www.dc.fi.udc.es/ai/medtool.html. 1997.
J. Mira. Computación neuronal. Curso de la UNED, 1992.
J. Mira and A.E. Delgado. Linear and algorithmic formulation of cooperative computation in neural nets. In F. Pichler and Moreno-Diaz, editors, Computer Aided Systems Theory, pages 2–20. Springer-Verlag, 1991. Lecture Notes in Computer Science, Vol. 585.
T. Nordstrom and B. Svensson. Using and designing massively parallel computers for artificial neural networks. J. Parallel and Distributed Computing, 14(3):260–285, 1992.
J. Santos, R.P. Otero, and J. Mira. Nettool: A hydrid connectionist-symbolic development environment. In J. Mira and F. Sandoval, editors, From Natural, to Artificial Neural Computation, pages 658–665. Springer-Verlag, 1995. Lecture Notes in Computer Science, Vol. 930.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Santos, J., Cabarcos, M., Otero, R.P., Mira, J. (1997). Parallelization of connectionist models based on a symbolic formalism. In: Mira, J., Moreno-Díaz, R., Cabestany, J. (eds) Biological and Artificial Computation: From Neuroscience to Technology. IWANN 1997. Lecture Notes in Computer Science, vol 1240. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0032488
Download citation
DOI: https://doi.org/10.1007/BFb0032488
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63047-0
Online ISBN: 978-3-540-69074-0
eBook Packages: Springer Book Archive