Applying hopfield neural networks to solve CSP problems

Balanda, Anatolii; Микола Погорецький; Serhieieva, Diana; Hribov, Mikhail; Toporetska, Zoriana

doi:10.30534/ijatcse/2020/190942020

Applying hopfield neural networks to solve CSP problems

Тип публікації :

Стаття

Дата випуску :

2020

Автор(и) :

Balanda, Anatolii

Academy named after Yevheniy Bereznyak

Микола Погорецький

Кафедра кримінального процесу та криміналістики

Serhieieva, Diana

Taras Shevchenko National University of Kyiv

Hribov, Mikhail

National Academy of Internal Affairs of Kyiv

Toporetska, Zoriana

Taras Shevchenko National University of Kyiv

Мова основного тексту :

Англійська

eKNUTSHIR URL :

https://ir.library.knu.ua/handle/15071834/8332

DOI :

10.30534/ijatcse/2020/190942020

Журнал :

International Journal of Advanced Trends in Computer Science and Engineering

Том :

9

Випуск :

4

ISSN :

2278-3091

Початкова сторінка :

4724

Кінцева сторінка :

4728

Цитування :

[APA 7] Balanda, A., Микола, П., Serhieieva, D., Hribov, M., & Toporetska, Z. (2020). Applying hopfield neural networks to solve CSP problems. International Journal of Advanced Trends in Computer Science and Engineering, 9(4), 4724–4728. https://doi.org/10.30534/ijatcse/2020/190942020

[ДСТУ] Applying hopfield neural networks to solve CSP problems / A. Balanda et al. International Journal of Advanced Trends in Computer Science and Engineering. 2020. Vol. 9, no. 4. P. 4724—4728. DOI: 10.30534/ijatcse/2020/190942020 (date of access: 25.07.2026).

The article reviews methods based on the Hopfield neural network for solving CSP and FCSP problems. The first attempt to apply this type of neural network to solving the CSP problem was made by Hopfield himself, after which a number of modifications of the original algorithm took place. That is, all the methods presented in the article are modifications of each other and have developed consistently. Some characteristics of Hopfield network-based methods in comparison with other (non-neural network-based) algorithms and CSP solutions are also given. In the field of artificial intelligence, there is a class of combinatorial problems called CSP problems (Constraint satisfaction Problems). They are a powerful tool for solving practical problems that can be designed for many variables that are bound together by constraints.

Ключові слова :

CSP constraint satisfaction problem neural network Hopfield neural network

Галузі знань та спеціальності :

08 Право

Галузі науки і техніки (FOS) :

Право

Файл(и) :

Завантажити

Формат :

Adobe PDF

Розмір :

397.18 KB

Контрольна сума :

(MD5):e0d41542793e83eb3edbd866182abf33

Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International

Якщо не вказано інше, ця робота розповсюджується на умовах ліцензії Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International