Geodesic
The architecture of a~neural network with a~sequential division of images into pairs
Prikladnaâ diskretnaâ matematika, no. 3 (2018), pp. 98-109.

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

The known classical architectures of neural networks have many weak properties and high limitations such as great difficulties in choosing proper parameters (the numbers of neurons, connections, layers), in learning and in expanding a learned network and some others. In this paper, to overcome these shortcomings, we consider the architecture of a neural network in which the recognition is performed in pairs. The neural network is created on the basis of the neural network architecture that implements the method of the nearest neighbor, without using analytical expressions and a set of selected samples. The paper shows that such an architecture can be applied to the recognition problem with a very large number of images (classes). At the same time, the proposed neural network has a simple architecture, with the possibility of simpler learning of the neural network. The neural network architecture allows to add new recognizable images to the neural network without changing the previous network settings.
Keywords: neural network architectures, neural networks, neurocomputer, convolutional networks, neural network training algorithms. 
@article{PDM_2018_3_a9,
     author = {P. Sh. Geidarov},
     title = {The architecture of a~neural network with a~sequential division of images into pairs},
     journal = {Prikladna\^a diskretna\^a matematika},
     pages = {98--109},
     publisher = {mathdoc},
     number = {3},
     year = {2018},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/PDM_2018_3_a9/}
}
TY  - JOUR
AU  - P. Sh. Geidarov
TI  - The architecture of a~neural network with a~sequential division of images into pairs
JO  - Prikladnaâ diskretnaâ matematika
PY  - 2018
SP  - 98
EP  - 109
IS  - 3
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/PDM_2018_3_a9/
LA  - ru
ID  - PDM_2018_3_a9
ER  - 
%0 Journal Article
%A P. Sh. Geidarov
%T The architecture of a~neural network with a~sequential division of images into pairs
%J Prikladnaâ diskretnaâ matematika
%D 2018
%P 98-109
%N 3
%I mathdoc
%U http://geodesic.mathdoc.fr/item/PDM_2018_3_a9/
%G ru
%F PDM_2018_3_a9
P. Sh. Geidarov. The architecture of a~neural network with a~sequential division of images into pairs. Prikladnaâ diskretnaâ matematika, no. 3 (2018), pp. 98-109. http://geodesic.mathdoc.fr/item/PDM_2018_3_a9/

[1] Barskiy A. B., Neural Networks: Recognition, Management, Decision Making, Finansy i Statistika Publ., Moscow, 2004 (in Russian)

[2] Golovko V. L., Neural Networks: Training, Organization and Application, Book 4, IPRJR Publ., Moscow, 2001 (in Russian)

[3] Wasserman F., Neurocomputing. Theory and Practice, Mir Publ., Moscow, 1992 (in Russian)

[4] LeCun Y., Bengio Y., Hinton G., “Deep learning”, Nature, 521 (2015), 436–444 | DOI

[5] Schmidhuber J., “Deep learning in neural networks: An overview”, Neural Networks, 61 (2015), 85–117 | DOI

[6] Neural Networks and Deep Learning, http://www.neuralnetworksanddeeplearning.com/index.html

[7] Geidarov P. Sh., “Neural networks on the basis of the sample method”, Automatic Control and Computer Sci., 43:4 (2009), 203–210 | DOI

[8] Geidarov P. Sh., “Multitasking application of neural networks implementing metric methods of recognition”, Autom. Remote Control, 74:9 (2013), 1474–1485 | DOI | Zbl

[9] Birger I. A., Technical Diagnostics, Mashinostroenie Publ., Moscow, 1978 (in Russian)

[10] Geidarov P. Sh., “Algorithm for realizing the method of the nearest neighbor in a multilayer perceptron”, Trudy SPIIRAN, 51 (2017), 123–151 (in Russian) | DOI

[11] Vasin D. Yu., Aratskiy A. V., “Character recognition based on invariant moments of graphic images”, 25th Intern. GraphiCon Conf., Moscow, 2015, 259–264 (in Russian)