This Article 
 Bibliographic References 
 Add to: 
A Framework for Neural Net Specification
July 1992 (vol. 18 no. 7)
pp. 601-612

A notation for the specification of neural nets is proposed. The aim is to produce a simple mathematical framework for use in specifying neural nets essentially by defining their transfer functions and connections. Nets are specified as interacting processing elements (nodes), communicating via instant links. Dynamics and adaptation are defined at the processing elements themselves, and all interaction is explicitly specified by directed arcs. Specifications can be built up hierarchically by turning a specification into a generator for a node, or they can be developed top-down. The use of the system is illustrated.

[1] C. Hewitt, "Viewing control structures as patterns of passing messages,"J. of Artificial Intelligence, vol. 8, June 1977.
[2] B. Derotet al., "NACRE: A neuron-oriented programming environment," inNeuro-Nimes '89, 1989.
[3] P.C. Treleaven, "PYGMALION: Neural network programming environment," in T. Kohonenet al., Eds.,Artificial Neural Networks. North-Holland, 1991.
[4] ANSpecTMUser's Manual, Science Applications International Corporation, San Diego, CA, 1989.
[5] R. Hecht-Nielson,Neurocomputing. Reading, MA: Addison-Wesley, 1990.
[6] P. Bessiereet al., MENTAL: A Virtual Machine Approach to Artificial Neural Networks Programming, ESPIRIT BRA Project 3049 Final Report, June 30, 1991.
[7] P. Koikkalainen, "MIND: A specification formalism for neural networks," in T. Kohonenet al., Eds.,Artificial Neural Networks, North-Holland, 1991.
[8] C. A. R. Hoare, "Communicating sequential processes,"Commun. ACM, vol. 21, pp. 666-677, 1978.
[9] R. Milner, "Flowgraphs and flow algebras,"J. ACM, vol. 26, 1979.
[10] O. Ekeberget al., "SWIM--A simulator for real neural networks,"Royal Institute of Technology Studies in Artificial Neural Systems TRITA-NA- P9014, 1990.
[11] N. T. Carnevaleet al., "Neuron simulations with Saber,"J. Neuroscience Methods, vol. 33, 1990.
[12] A. Vladimirescuet al., "SPICE Version 2G Users Guide," Univ. California, Aug. 1981.
[13] P. Caspi, D. Pilaud, N. Halbwachs, and J. Plaice, LUSTRE: A declarative language for programming synchronous systems," in14th ACM Symposium on Principles of Programming Languages, Jan. 1987.
[14] E. Gelenbe, "Random neural networks with negative and positive signals and product form solution,"Neural Computation, vol. 1, pp. 502-510, 1989.
[15] D.E. Rumelhartet al., "Learning representations by backpropagating errors,"Nature, Oct. 1986.
[16] R.J. Williams and D. Zipser, "A learning algorithm for continually running fully recurrent neural networks,Neural Computation, vol. 1, pp. 270-280, 1989.
[17] T.D. Sanger, "Optimal unsupervised learning in a single layer feedforward neural network,"Neural Networks, vol. 2, pp. 459-473, 1989.
[18] A.W. Thomson, "Modeling neural nets," M.Sc thesis, Univ. Stirling Department of Computing Science, Apr. 1990.
[19] L.S. Smith, "Formalizing neural networks," inNeural Networks from Models to Applications, in L. Personnazet al., Eds. Paris: IDSET, 1989.

Index Terms:
framework; neural net specification; mathematical framework; transfer functions; connections; interacting processing elements; formal specification; neural nets
L.S. Smith, "A Framework for Neural Net Specification," IEEE Transactions on Software Engineering, vol. 18, no. 7, pp. 601-612, July 1992, doi:10.1109/32.148478
Usage of this product signifies your acceptance of the Terms of Use.