Abstract
In this paper we investigate how a neural network can be regarded as a massive parallel computer architecture. To this end, we focus our attention not on a stochastic asynchronous model, but on the McCulloch and Pitts network [MP43], as modified by Caianiello [Cai61], and we specify what we mean for the environment in which the network operates: it is essentially the entity assigning meaning to the network input and output nodes. Changing the environment definition implies dealing with different neural architectures.
To show how to program a neural architecture, we introduce a model of environment helping us in choosing functions suitable to be pipelined, as in a data flow architecture. As an example, we sketch the working of a parallel multiplier function.
This is a preview of subscription content, log in via an institution.
Preview
Unable to display preview. Download preview PDF.
References
E. R. Caianiello. Outline of a theory of thought processes and thinking machines. J. of Theor. Biol., (2):204–235, 1961.
J. B. Dennis. Data flow supercomputers. Computer, 48–56, november 1980.
P. De Pinto. The implementation of the assembler associated to a general purpose neural network. In Third Italian Workshop on “Parallel Architectures and Neural Networks”, Vietri sul Mare (SA), May 1990.
P. De Pinto, F. E. Lauria, and M. Sette. On the Hebb rule and unlimited precision arithmetic in a MacCulloch and Pitts network. In R. Eckmiller, editor, Advanced neural computers, pages 121–128, International Symposium on: Neural Networks for Sensory and Motor Systems, Elsevier Sci., Amsterdam, 1990.
C. Hewitt. The challenge of open systems. BYTE, 223–242, April 1985.
J. E. Hopcroft and J. D. Ullman. Introduction to automata theory, languages and computation. Addison-Wesley Pub. Co., 1979.
F. E. Lauria. A connectionist approach to knowledge acquisition. Cybernetics and Systems: an Intern. Journ., (5):19–35, 1988.
F. E. Lauria. A general purpose neural network as a new computing paradigm. In E. R. Caianiello, editor, Second Italian Workshop on “Parallel Architectures and Neural Networks”, pages 131–144, World Scientific Pub., Singapore, 1989.
W. S. McCulloch and W. Pitts. A logical calculus of the ideas immanent in the nervous activity. Bull. Math. Bioph., 115–143, 1943.
D. E. Rumelhart and J. L. McClelland. Parallel distributed processing. Volume 1, MIT Press, Cambridge, MA, 1986.
M. Sette. An OCCAM simulation of a general purpose neural network. In Third Italian Workshop on “Parallel Architectures and Neural Networks”, Vietri sul Mare (SA), May 1990.
G. Trautteur. Problems with Symbols. A commentary to: Herbert Simon, “Scientific Discovery as Problem Solving”. RISESST, (1), 1991. To be published.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1991 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
De Pinto, P., Sette, M. (1991). On neural network programming. In: Ardizzone, E., Gaglio, S., Sorbello, F. (eds) Trends in Artificial Intelligence. AI*IA 1991. Lecture Notes in Computer Science, vol 549. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-54712-6_254
Download citation
DOI: https://doi.org/10.1007/3-540-54712-6_254
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-54712-9
Online ISBN: 978-3-540-46443-3
eBook Packages: Springer Book Archive