Geodesic
The uplearning of the Hopfield neural network: the search for a global minimum of a functional and the model of rapid eye movement
Matematičeskoe modelirovanie, Tome 21 (2009) no. 5, pp. 10-20 
A. A. Ezhov; A. S. Cherepnev. The uplearning of the Hopfield neural network: the search for a global minimum of a functional and the model of rapid eye movement. Matematičeskoe modelirovanie, Tome 21 (2009) no. 5, pp. 10-20. http://geodesic.mathdoc.fr/item/MM_2009_21_5_a1/
@article{MM_2009_21_5_a1,
     author = {A. A. Ezhov and A. S. Cherepnev},
     title = {The uplearning of the {Hopfield} neural network: the search for a~global minimum of a~functional and the model of rapid eye movement},
     journal = {Matemati\v{c}eskoe modelirovanie},
     pages = {10--20},
     year = {2009},
     volume = {21},
     number = {5},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/MM_2009_21_5_a1/}
}
TY  - JOUR
AU  - A. A. Ezhov
AU  - A. S. Cherepnev
TI  - The uplearning of the Hopfield neural network: the search for a global minimum of a functional and the model of rapid eye movement
JO  - Matematičeskoe modelirovanie
PY  - 2009
SP  - 10
EP  - 20
VL  - 21
IS  - 5
UR  - http://geodesic.mathdoc.fr/item/MM_2009_21_5_a1/
LA  - ru
ID  - MM_2009_21_5_a1
ER  - 
%0 Journal Article
%A A. A. Ezhov
%A A. S. Cherepnev
%T The uplearning of the Hopfield neural network: the search for a global minimum of a functional and the model of rapid eye movement
%J Matematičeskoe modelirovanie
%D 2009
%P 10-20
%V 21
%N 5
%U http://geodesic.mathdoc.fr/item/MM_2009_21_5_a1/
%G ru
%F MM_2009_21_5_a1

Voir la notice de l'article provenant de la source Math-Net.Ru

The traveling salesman problem is solved using the Hopfield neural network model. The solution comes to the energy minimization of the neural network for the purpose of finding the global minimum of an appropriate functional. The procedure “Uplearning” is introduced. In addition, local minima become deeper rising one's accessibility in this process. In this case, just the global minimum survives frequently. The results are represented which help to guess that the Uplearning increases the probability to find the global minimum in contrast to the Unlearning which was introduced by Hopfield in 1983.

[1] Pierluigi Crescenzi, Viggo Kann, A compendium of NP optimization problems

[2] David S. Johnson, Lyle A. McGeoch, The Traveling Salesman Problem: A Case Study in Local Optimization, November 20, 1995

[3] Bernd Fritzke, Peter Wilke, FLEXMAP – A Neural Network For The Traveling Salesman Problem With Linear Time And Space Complexity, Germany, 1995

[4] J. J. Hopfield, “Neural Networks and Physical Systems with Emergent Collective Computational Abilities”, Proc. Natl. Acad. Sci. USA, 79 (1982), 2554 | DOI | MR

[5] W. S. McCulloch, W. A. Pitts, Bull. Math. Biophys., 5 (1969), 115 | DOI | MR

[6] D. O. Hebb, The Organization of Behaviour, Wiley, New York, 1949

[7] M. Minsky, S. Papert, Perceptrons, MIT Press, Cambridge, MA, 1969 | Zbl

[8] J. J. Hopfield, D. W. Tank, “Neural” Computation of Decisions in Optimization Problems, USA, 1985 | MR | Zbl

[9] J. J. Hopfield, D. L. Feinstein, R. G. Palmer, “ ‘Unlearning’ has a stabilizing effect in collective memories”, Nature, 304 (1983), 158–159 | DOI

[10] F. Crick, G. Mitchison, “The function of dream sleep”, Nature, 304 (1983), 111 | DOI

[11] K. A. Valiev, A. A. Kokin, Kvantovye kompyutery: nadezhdy i realnost, RKhD, Izhevsk, 2004

[12] A. A. Ezhov, Yu. A. Kalambef, L. A. Knizhnikova, “Neural networks: general properties particular applications”, Neural Networks: Theory and Architectures, eds. V. I. Kryukov, A. Holden, Manchester University Press, Manchester, UK, 1990, 39–47 | MR

[13] A. A. Ezhov, V. L. Vvedensky, “Object generation with neural networks (when spurious memories are useful)”, Neural Networks, 9 (1996), 1491–1495 | DOI | Zbl