Библиографическое описание:An Efficient Training Algorithm of Restricted Boltzmann Machines : доклад, тезисы доклада / V. V. Matskevich, V. A. Stasiuk. - [S. l. : s. n.], 2022. - Текст : непосредственный // HYBRID METHODS OF MODELING AND OPTIMIZATION IN COMPLEX SYSTEMS : Proceedings of the International Workshop “Hybrid methods of modeling and optimization in complex systems” (HMMOCS 2022) / Krasnoyarsk Regional Science and Technology City Hall, Russia Siberian Federal University, Reshetnev Siberian State University of Science and Technology ; International Workshop “Hybrid methods of modeling and optimization in complex systems” (HMMOCS 2022) (2022 ; 22.11 - 24.11 ; Krasnoyarsk). - London, United Kingdom : European Proceedings, 2022. - P. 296-303. - ISBN 9781802969603, DOI 10.15405/epct.23021.36.
Аннотация:The paper deals with an actual applied problem related to the artificial neural networks training. An approach to the solution based on the idea of random search is proposed. An original training algorithm that implements Boltzmann annealing has been developed and its convergence in probability to the global optimum has been proved. It is also shown that the proposed algorithm can be easily modified to train any artificial neural network. Thus, it has a good prospect for solving applied problems using neural network technologies in general. Experimental studies have been carried out, in which, using the example of compressing color raster images problem, the proposed algorithm was compared with the known adaptive moment algorithm - one of the best gradient methods for training neural networks. Image compression was performed using an ensemble of n Gauss-Bernoulli restricted Boltzmann machines. The use of an ensemble of n machines in combination with a specially developed parallelization procedure made it possible to reduce the computational complexity of the training process and increase the speed of the proposed algorithm. As a result of experiments, it was shown that the proposed approach is not inferior to gradient methods in terms of speed. Moreover, the developed training algorithm turned out to be more than twice as effective as the adaptive moment algorithm in terms of the quality of the solution obtained.
