 |
advertisement |
|
|
|
|
|
|
Instruments and Systems: Monitoring, Control, and Diagnostics Annotation << Back
Evaluation of Kolmogorov-Arnold Networks Effi
ciency for Fire Robot Control Tasks |
I.N. POZHARKOVA
The article is devoted to the study of the effectiveness of using Kolmogorov-Arnold networks in the tasks of controlling fi re robots, in
particular, predicting the geometric characteristics and trajectories of free jets of a fi re extinguishing substance. Comparative analysis
technology of models on the basis of new architecture and multilayer perceptron with appropriate calculations is described. Principles of
network structure construction, formation of training and test samples, used training strategies and optimization methods are given. The
results of comparative analysis with assessment of learning speed, size of built models, their speed and accuracy are presented.
Keywords: fi re robots, free jets prediction, machine learning, Kolmogorov-Arnold network, multilayer perceptron, neural network, KAN,
MLP.
Pp. 33-41. |
|
|
|
Last news:
Выставки по автоматизации и электронике «ПТА-Урал 2018» и «Электроника-Урал 2018» состоятся в Екатеринбурге Открыта электронная регистрация на выставку Дефектоскопия / NDT St. Petersburg Открыта регистрация на 9-ю Международную научно-практическую конференцию «Строительство и ремонт скважин — 2018» ExpoElectronica и ElectronTechExpo 2018: рост площади экспозиции на 19% и новые формы контент-программы Тематика и состав экспозиции РЭП на выставке "ChipEXPO - 2018" |