Matches in SemOpenAlex for { <https://semopenalex.org/work/W3201391500> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W3201391500 endingPage "369" @default.
- W3201391500 startingPage "360" @default.
- W3201391500 abstract "In the process of conducting various physical experiments, a certain set of data is accumulated. Processing and interpreting the simulation results is a fundamental task for analyzing the behaviour of the model, predicting its future actions, and managing the entire system. This paper provides an overview of the currently existing approaches to identification of dynamic systems models: white, gray and black boxes. Special attention is paid to methods based on neural networks. The article suggests a combined approach that allows both preserving a physical consistency of the model and using modern methods for learning from data. A polynomial neural network of a special architecture, approximating the general solution of the system of ordinary differential equations (ODEs) in the form of Taylor map is considered. This model can work in the case of a small amount of initial data, which is a problem when exploiting traditional machine learning methods, and neural networks in particular. The paper presents a new learning approach for PNN based on two steps: reconstructing an ODEs system based on a single trajectory, and identifying a general solution to initialize the weights of a neural network. Neural network representation allows using traditional learning algorithms for additional fine-turning the weights in line with new measured data. A toy example of a nonlinear deflector demonstrates the power and generalization ability of the proposed algorithm." @default.
- W3201391500 created "2021-09-27" @default.
- W3201391500 creator A5031501826 @default.
- W3201391500 creator A5047024716 @default.
- W3201391500 creator A5081898677 @default.
- W3201391500 date "2021-01-01" @default.
- W3201391500 modified "2023-09-25" @default.
- W3201391500 title "Reconstruction and Identification of Dynamical Systems Based on Taylor Maps" @default.
- W3201391500 cites W1989989031 @default.
- W3201391500 cites W2038910178 @default.
- W3201391500 cites W2962872206 @default.
- W3201391500 cites W2963121266 @default.
- W3201391500 cites W2963363740 @default.
- W3201391500 cites W2969708557 @default.
- W3201391500 cites W2982585804 @default.
- W3201391500 doi "https://doi.org/10.1007/978-3-030-87010-2_26" @default.
- W3201391500 hasPublicationYear "2021" @default.
- W3201391500 type Work @default.
- W3201391500 sameAs 3201391500 @default.
- W3201391500 citedByCount "1" @default.
- W3201391500 countsByYear W32013915002022 @default.
- W3201391500 crossrefType "book-chapter" @default.
- W3201391500 hasAuthorship W3201391500A5031501826 @default.
- W3201391500 hasAuthorship W3201391500A5047024716 @default.
- W3201391500 hasAuthorship W3201391500A5081898677 @default.
- W3201391500 hasConcept C11413529 @default.
- W3201391500 hasConcept C116834253 @default.
- W3201391500 hasConcept C119247159 @default.
- W3201391500 hasConcept C119857082 @default.
- W3201391500 hasConcept C124101348 @default.
- W3201391500 hasConcept C134306372 @default.
- W3201391500 hasConcept C154945302 @default.
- W3201391500 hasConcept C177148314 @default.
- W3201391500 hasConcept C2780009758 @default.
- W3201391500 hasConcept C28826006 @default.
- W3201391500 hasConcept C33923547 @default.
- W3201391500 hasConcept C34862557 @default.
- W3201391500 hasConcept C41008148 @default.
- W3201391500 hasConcept C50644808 @default.
- W3201391500 hasConcept C51544822 @default.
- W3201391500 hasConcept C59822182 @default.
- W3201391500 hasConcept C78045399 @default.
- W3201391500 hasConcept C86803240 @default.
- W3201391500 hasConceptScore W3201391500C11413529 @default.
- W3201391500 hasConceptScore W3201391500C116834253 @default.
- W3201391500 hasConceptScore W3201391500C119247159 @default.
- W3201391500 hasConceptScore W3201391500C119857082 @default.
- W3201391500 hasConceptScore W3201391500C124101348 @default.
- W3201391500 hasConceptScore W3201391500C134306372 @default.
- W3201391500 hasConceptScore W3201391500C154945302 @default.
- W3201391500 hasConceptScore W3201391500C177148314 @default.
- W3201391500 hasConceptScore W3201391500C2780009758 @default.
- W3201391500 hasConceptScore W3201391500C28826006 @default.
- W3201391500 hasConceptScore W3201391500C33923547 @default.
- W3201391500 hasConceptScore W3201391500C34862557 @default.
- W3201391500 hasConceptScore W3201391500C41008148 @default.
- W3201391500 hasConceptScore W3201391500C50644808 @default.
- W3201391500 hasConceptScore W3201391500C51544822 @default.
- W3201391500 hasConceptScore W3201391500C59822182 @default.
- W3201391500 hasConceptScore W3201391500C78045399 @default.
- W3201391500 hasConceptScore W3201391500C86803240 @default.
- W3201391500 hasLocation W32013915001 @default.
- W3201391500 hasOpenAccess W3201391500 @default.
- W3201391500 hasPrimaryLocation W32013915001 @default.
- W3201391500 hasRelatedWork W10555355 @default.
- W3201391500 hasRelatedWork W12091112 @default.
- W3201391500 hasRelatedWork W1243554 @default.
- W3201391500 hasRelatedWork W12940665 @default.
- W3201391500 hasRelatedWork W14993799 @default.
- W3201391500 hasRelatedWork W2008406 @default.
- W3201391500 hasRelatedWork W206613 @default.
- W3201391500 hasRelatedWork W5641948 @default.
- W3201391500 hasRelatedWork W700355 @default.
- W3201391500 hasRelatedWork W9364628 @default.
- W3201391500 isParatext "false" @default.
- W3201391500 isRetracted "false" @default.
- W3201391500 magId "3201391500" @default.
- W3201391500 workType "book-chapter" @default.