Please use this identifier to cite or link to this item:
https://elib.bsu.by/handle/123456789/291836| Title: | Random search in neural networks training |
| Authors: | Krasnoproshin, V. V. Matskevich, V. V. |
| Keywords: | ЭБ БГУ::ЕСТЕСТВЕННЫЕ И ТОЧНЫЕ НАУКИ::Математика ЭБ БГУ::ЕСТЕСТВЕННЫЕ И ТОЧНЫЕ НАУКИ::Кибернетика |
| Issue Date: | 2022 |
| Publisher: | Minsk : BSU |
| Citation: | Computer Data Analysis and Modeling: Stochastics and Data Science : Proc. of the XIII Intern. Conf., Minsk, Sept. 6–10, 2022 / Belarusian State University ; eds.: Yu. Kharin [et al.]. – Minsk : BSU, 2022. – Pp. 96-99. |
| Abstract: | At the present time digital technologies are impetuous developing. In frames of this process neural networks take important place. Based on them, in particular, many intellectual decision support systems in medicine, monitoring of environment etc. are built. Wide spread of neural networks is conditioned by their universality. Unlike classical data processing algorithms they can fast adapt to solving a specific application problem [3]. Adaptation process is called neural network training. It plays a key role in quality of obtained solution. With grows of processed data volume neural network tunable parameters count grows. That’s why neural networks training problem is still actual [2]. This paper proposes an original neural networks training algorithm, based on random search idea |
| URI: | https://elib.bsu.by/handle/123456789/291836 |
| ISBN: | 978-985-881-420-5 |
| Licence: | info:eu-repo/semantics/restrictedAccess |
| Appears in Collections: | 2022. Computer Data Analysis and Modeling: Stochastics and Data Science |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.