Management decisions based on reinforcement learning method
https://doi.org/10.28995/2073-6304-2024-1-38-48
Abstract
The article considers the issues of evaluating control decisions for repetitive tasks. It is shown that using the reinforcement learning method makes it possible to increase the objectivity of assessments of possible management decisions. Authors highlight that the proposed method has a methodological connection with the game-theoretic approach and can be generalized for various options in practical application.
About the Authors
T. N. Borovik
MIREA – Russian Technological University
Russian Federation
Russian Federation
Tat’yana N. Borovik
bld. 78, Vernadskii Avenue, Moscow, 119454
R. V. Shamin
MIREA – Russian Technological University
Russian Federation
Russian Federation
Roman V. Shamin, Dr. of Sci. (Physics and Mathematics)
bld. 78, Vernadskii Avenue, Moscow, 119454
References
1. Сormen, T., Leiserson, Ch., Rivest, R. and Stein, K. (2005), Algoritmy: postroenie i analiz [Algorithms. Construction and analysis = Introduction to Algorithms], Williams, Moscow, Russia.
2. Neumann, J. von and Morgenstern, O. (1970), Teoriya igr i ekonomicheskoe povedenie [Theory of Games and economic behavior], Nauka, Moscow, Russia.
3. Nikolenko, S.I. and Tulupyev, A.L. (2009), Self-learning systems, MTsNMO, Moscow, Russia.
4. Sutton, R.S. and Barto, E.G. (2020), Obuchenie s podkrepleniem [Reinforcement Learning], DMK press, Moscow, Russia.
5. Winder, Ph. (2023), Obuchenie s podkrepleniem dlya real’nykh zadach. Inzhenernyi podkhod [Reinforcement learning for actual tasks. Engineering approach], BKhV-Peterburg, Moscow, Russia.
Review
For citations:
Borovik T.N., Shamin R.V. Management decisions based on reinforcement learning method. RSUH/RGGU BULLETIN. Series Economics. Management. Law. 2024;(1):38-48. (In Russ.) https://doi.org/10.28995/2073-6304-2024-1-38-48
Views:
223
Email this article 