Abstract
The problem of improving structural properties of artificial discrete neural networks is investigated. The structure of the neural network is regarded as a theoretical graph. Cyclical substructures can occur in this structure under certain conditions, e.g., when there is feedback among neural network layers. Some properties of cycles in the graphs corresponding to neural networks have a significant effect not only on the rate of convergence to the (stable) solutions of the problems posed by network users, but also on the very possibility of obtaining these solutions. These properties include the negativity of some circuits (simple cycles) in neural network graphs. We propose an algorithm that makes it possible to eliminate negative circuits from neural network graphs under certain constraints formulated in this paper. It increases the chances of finding correct solutions to the problems for which neural networks were developed. An illustrative example is presented.
Similar content being viewed by others
REFERENCES
Cireşan, D., Meier, U., Masci, J., and Schmidhuber, J., Multi-column deep neural network for traffic sign classification, Neural Networks, 2012.
Talbot, D., CES 2015: Nvidia demos a car computer trained with “deep learning,” MIT Technol. Rev., 2015.
Roth, S., Shrinkage fields for effective image restoration, Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2014.
Deng, L. and Yu, D., Deep learning: Methods and applications, Found. Trends Signal Process., 2014, vol. 7, nos. 3–4, pp. 1–19.
Rosenblatt, F., Principles of Neurodynamics: Perceptrons and the Theory of Brain Mechanisms, Washington: Spartan Books, 1961.
Parallel Distributed Processing: Explorations in the Microstructure of Cognition. Volume 1: Foundation, Rumelhart, D.E., McClelland, J.L., and PDP research group, Eds., MIT Press, 1986.
Hopfield, J.J., Neural networks and physical systems with emergent collective computational abilities, Proc. Natl. Acad. Sci. USA, 1982, vol. 79, no. 8, pp. 2554–2558.
Ackley, D.H., Hinton, G.E., and Sejnowski, T.J., A learning algorithm for Boltzmann machines, Cognit. Sci., 1985, vol. 9, no. 1, pp. 147–169.
Kohonen, T., Self-organized formation of topologically correct feature maps, Biol. Cybern., 1982, vol. 43, no. 1, pp. 59–69.
Elman, J.L., Finding structure in time, Cognit. Sci., 1990, vol. 14, no. 2, pp. 79–211.
Lecun, Y., Bottou, L., Bengio, Y., and Haffner, P., Gradient-based learning applied to document recognition, Proc. IEEE, 1998, vol. 86, no. 11, pp. 2278–2324.
Graves, A., Greg, W., and Ivo, D., Neural Turing machines, 2014.
Ritter, G.X. and Sussner, P., An introduction to morphological neural networks, Proc. 13th Int. Conf. Pattern Recognition, Vienna, 1996, pp. 709–717.
Karpov, Yu.L., Karpov, L.E., and Smetanin, Yu.G., Adaptation of general concepts of software testing to neural networks, Program. Comput. Software, 2018, vol. 44, no. 5, pp. 324–334.
IEEE 829 Standard for Software Test Documentation and IEEE 1008 Standard for Software Unit Testing. https://www.twirpx.com/file/1615980.
ISO/MEK 12119. Information technology. Software packages. Quality requirements and testing. http://docs.cntd.ru/document/1200025075.
GOST R 56920-2016, GOST R 56921–2016, and GOST R 56922–2016. https://allgosts.ru.
ISO/IEC 29119-2013 1-5. Software testing. http://files.stroyinf.ru/Data2/1/4293754/ 4293754866.pdf.
GOST R 12207-2010 and ISO/IEC 12207:2008. http://docs.cntd.ru/document/1200082859.
Aracena, J., Demongeot, J.D., and Goles, E., Positive and negative circuits in discrete neural networks, IEEE Trans. Neural Networks, 2004, vol. 15, no. 1.
Prim, R.C., Shortest connection networks and some generalizations, Bell Syst. Tech. J., 1957, vol. 36, no. 6, pp. 1389–1401.
Cormen, T.H., Charles, E., Leiserson, C.E., Rivest, R.L., and Stein, C., Introduction to Algorithms, MIT Press, McGraw-Hill, 2002, 2nd ed.
Funding
This work was supported by the Russian Foundation for Basic Research, project nos. 18-07-00697-а, 18-07-01211-а, 19-07-00321-а, and 19-07-00493-а.
Author information
Authors and Affiliations
Corresponding authors
Additional information
Translated by Yu. Kornienko
Rights and permissions
About this article
Cite this article
Karpov, Y.L., Karpov, L.E. & Smetanin, Y.G. Elimination of Negative Circuits in Certain Neural Network Structures to Achieve Stable Solutions. Program Comput Soft 45, 241–250 (2019). https://doi.org/10.1134/S0361768819050025
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0361768819050025