Analysis of the possibilities of self-adaptive neural networks with search behavior in agro-ecological systems
DOI:
https://doi.org/10.34121/1028-9763-2020-3-125-133Keywords:
analysis, capabilities, neuroinformatics, self-adaptation, neural networks, search behavior, agroecological systems, neuroinformation control systems, аналіз, можливості, нейроінформатика, самоадаптація, нейронні мережі, пошукова поведінка, агроекологічні системи, нейроінформаційна система управлінняAbstract
Abstract. The complexity of environmental problems facing modern science in connection with the deterioration of the environmental situation on the planet and the growing dynamics of ongoing processes are constantly growing (this primarily concerns agro-ecological systems). At the same time, the flexibility and accuracy of ecological models created by traditional mathematical methods, as well as the speed of their construction, in practice often does not live up to expectations. The same can be said about the engineering management of agro-ecological facilities. Among the most flexible and effective ways to solve such problems, neural network models and neurocomputers are highlighted. However, the concepts underlying the construction of modern neural network training algorithms impose serious limitations on the potential range of application of neuroinformatics achievements in solving environmental problems of agricultural systems. The number of scientific publications with radically new results is steadily decreasing and existing developments are beginning to “spread” across applications. The manifestation of such trends indicates that the main potential of the ideas that caused the next progress in this most important bionic direction and the creation of the 6th generation of computers – neurocomputers are exhausted. Modern advances in neuroinformatics, based on the use of supervisor algorithms, are mainly associated with the possibility of using hidden layers of neurons (not connected to the input and output), which provided high adaptive capabilities of neural networks, and universality based on the ability to train a neural network to solve a precisely posed problem. In this paper, the main limitations inherent in modern approaches to limiting neural networks are indicated, and the concept of constructing a new type of training neural network and network algorithms is proposed. Some non-traditional opportunities provided by the proposed concept are described. The conceptual foundations of the development of the neuroinformation system for controlling the electrotechnical complex of the information and technical system for local operational monitoring are proposed.References
1. McCulloch W., Pitts W. А logical calculus of the ideal imminent in nervous activity. Bull. Math.Bioplis. 1943. Vol. S. P. 115–137.
2. Rosenblatt F. Principles of neurodynamics. Baltimore, 1962. 420 p.
3. Minsky M., Papert S. Perceptrons. Cambridge, Massachusets, 1969. 380 p.
4. Hopfild J.J. Neural networks and physical systems with emergent collective computational abilities. Proc. Nat. Acad. Sci. USA. 1982. Vol. 79. P. 2554–2558.
5. Барцев С.И., Охонин В.А. Адаптивные сети обработки информации. Красноярск, 1986. 19 с. (Препринт № 59Б, Институт биофизики СОАН СССР).
6. Bartsev S.I., Okhonin V.A. Variation principle and the algorithm of dual functioning: examples of applications. Neurocomputers and attention II: connectionism and neurocomputers. Manchester and New York, 1991. P. 445–452.
7. Rumelhart D.E., Hinton G.E., Williams R.J. Learning representatives by back-propagating errors. Nature. 1986. Vol. 323. P. 533–536.
8. Bartsev S.I., Okhonin V.A. A self-learning neural networks plain «two coins». Neurocomputers and attention II: connectionism and neurocomputers. Manchester and New York. 1991. P. 453–458.
9. Lankin J.P., Khlebopros R.G. Modeling of changes of Ecological objects with the Help of Neuron Networks. Institute of Biophysics SB RAC. Theor. Dept.; Reprint TO N2. Krasnoyarsk, 1997. 8 p.
10. Ланкин Ю.П. Самоадаптирующиеся нейронные сети. Красноярск, 1997. 21 с. (Препринт ТО № 3, Институт биофизики СО РАН. Теоротдел).
11. Ланкин Ю.П. Адаптивные сети с самостоятельной адаптацией. Красноярск, 1998. 17 с. (Препринт ТО № 4, Институт биофизики СО РАН. Теоротдел).
12. Ланкин Ю.П. Самоадаптирующиеся нейронные сети при решении экологических задач (возможности реализации поискового поведения). Инженерная экология. 1998. Т. 4. С. 2–11.
13. Охонин В.А., Смолянинова Л.Г., Абовский Н.П., Деруга А.П. Нейросетевое управление конструкциями и возможности физической реализации уравнений двойственного функционирования. Третий сибирский конгресс по прикладной и индустриальной математике, посвящѐнный памяти С.Л. Соболева (1908–1989): тезисы докладов. Новосибирск: изд-во Института математики СО РАН, 1998. T. V. С. 90.
14. Басканова Т.Ф. Алгоритмы самостоятельной адаптации для нейронных сетей. Красноярск, 1998. 14 с. (Препринт ТО № 5, Институт биофизики СО РАН. Теоротдел).
15. Kirkpatrick S., Gellat C.D., Vecchi M.D. Optimization by Stimulated Annealing. Science. 1983. Vol. 220. P. 671–680.
16. Widrow B. Adaptive Switching Circuits. IRE Wescow Couv. Record. 1960. Vol. 4. P. 96–104.
17. Лукашин Ю.П. Адаптивные методы краткосрочного прогнозирования. М.: Статистика, 1979. 252 с.
18. Троханяк В.І. Система енергозбереження у пташниках із використанням низькопотенціальної енергії ґрунту. К.: ЦП «Компрінт», 2018. 386 с.
Стаття надійшла до редакції 25.02.2020
135
84Published
Issue
Section
License
Copyright (c) 2020 Mathematical Machines and Systems
This work is licensed under a Creative Commons Attribution 4.0 International License.
