FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4381250521> ?p ?o ?g. }

• W4381250521 endingPage "61501" @default.
• W4381250521 startingPage "61481" @default.
• W4381250521 abstract "Many real-world systems can be described by mathematical models that are human-comprehensible, easy to analyze and help explain the system's behavior. Symbolic regression is a method that can automatically generate such models from data. Historically, symbolic regression has been predominantly realized by genetic programming, a method that evolves populations of candidate solutions that are subsequently modified by genetic operators crossover and mutation. However, this approach suffers from several deficiencies: it does not scale well with the number of variables and samples in the training data - models tend to grow in size and complexity without an adequate accuracy gain, and it is hard to fine-tune the model coefficients using just genetic operators. Recently, neural networks have been applied to learn the whole analytic model, i.e., its structure and the coefficients, using gradient-based optimization algorithms. This paper proposes a novel neural network-based symbolic regression method that constructs physically plausible models based on even very small training data sets and prior knowledge about the system. The method employs an adaptive weighting scheme to effectively deal with multiple loss function terms and an epoch-wise learning process to reduce the chance of getting stuck in poor local optima. Furthermore, we propose a parameter-free method for choosing the model with the best interpolation and extrapolation performance out of all the models generated throughout the whole learning process. We experimentally evaluate the approach on four test systems: the TurtleBot 2 mobile robot, the magnetic manipulation system, the equivalent resistance of two resistors in parallel, and the longitudinal force of the anti-lock braking system. The results clearly show the potential of the method to find parsimonious models that comply with the prior knowledge provided." @default.
• W4381250521 created "2023-06-20" @default.
• W4381250521 creator A5004892921 @default.
• W4381250521 creator A5033348430 @default.
• W4381250521 creator A5084264842 @default.
• W4381250521 date "2023-01-01" @default.
• W4381250521 modified "2023-10-18" @default.
• W4381250521 title "Toward Physically Plausible Data-Driven Models: A Novel Neural Network Approach to Symbolic Regression" @default.
• W4381250521 cites W1979769287 @default.
• W4381250521 cites W1991542861 @default.
• W4381250521 cites W2023140507 @default.
• W4381250521 cites W2170240482 @default.
• W4381250521 cites W2216139323 @default.
• W4381250521 cites W2516953263 @default.
• W4381250521 cites W2733964674 @default.
• W4381250521 cites W2783456932 @default.
• W4381250521 cites W2899283552 @default.
• W4381250521 cites W2954765142 @default.
• W4381250521 cites W2959065637 @default.
• W4381250521 cites W2963446712 @default.
• W4381250521 cites W2982316166 @default.
• W4381250521 cites W2991933648 @default.
• W4381250521 cites W3016401366 @default.
• W4381250521 cites W3033909669 @default.
• W4381250521 cites W3082465854 @default.
• W4381250521 cites W3106411828 @default.
• W4381250521 cites W3122808444 @default.
• W4381250521 cites W3150767555 @default.
• W4381250521 cites W3166656282 @default.
• W4381250521 cites W4310079394 @default.
• W4381250521 cites W4318344220 @default.
• W4381250521 doi "https://doi.org/10.1109/access.2023.3287397" @default.
• W4381250521 hasPublicationYear "2023" @default.
• W4381250521 type Work @default.
• W4381250521 citedByCount "0" @default.
• W4381250521 crossrefType "journal-article" @default.
• W4381250521 hasAuthorship W4381250521A5004892921 @default.
• W4381250521 hasAuthorship W4381250521A5033348430 @default.
• W4381250521 hasAuthorship W4381250521A5084264842 @default.
• W4381250521 hasBestOaLocation W43812505211 @default.
• W4381250521 hasConcept C104114177 @default.
• W4381250521 hasConcept C105795698 @default.
• W4381250521 hasConcept C110332635 @default.
• W4381250521 hasConcept C119857082 @default.
• W4381250521 hasConcept C122507166 @default.
• W4381250521 hasConcept C126838900 @default.
• W4381250521 hasConcept C132459708 @default.
• W4381250521 hasConcept C137800194 @default.
• W4381250521 hasConcept C154945302 @default.
• W4381250521 hasConcept C183115368 @default.
• W4381250521 hasConcept C2776400721 @default.
• W4381250521 hasConcept C33923547 @default.
• W4381250521 hasConcept C41008148 @default.
• W4381250521 hasConcept C50644808 @default.
• W4381250521 hasConcept C71924100 @default.
• W4381250521 hasConcept C8880873 @default.
• W4381250521 hasConceptScore W4381250521C104114177 @default.
• W4381250521 hasConceptScore W4381250521C105795698 @default.
• W4381250521 hasConceptScore W4381250521C110332635 @default.
• W4381250521 hasConceptScore W4381250521C119857082 @default.
• W4381250521 hasConceptScore W4381250521C122507166 @default.
• W4381250521 hasConceptScore W4381250521C126838900 @default.
• W4381250521 hasConceptScore W4381250521C132459708 @default.
• W4381250521 hasConceptScore W4381250521C137800194 @default.
• W4381250521 hasConceptScore W4381250521C154945302 @default.
• W4381250521 hasConceptScore W4381250521C183115368 @default.
• W4381250521 hasConceptScore W4381250521C2776400721 @default.
• W4381250521 hasConceptScore W4381250521C33923547 @default.
• W4381250521 hasConceptScore W4381250521C41008148 @default.
• W4381250521 hasConceptScore W4381250521C50644808 @default.
• W4381250521 hasConceptScore W4381250521C71924100 @default.
• W4381250521 hasConceptScore W4381250521C8880873 @default.
• W4381250521 hasFunder F4320335322 @default.
• W4381250521 hasLocation W43812505211 @default.
• W4381250521 hasLocation W43812505212 @default.
• W4381250521 hasLocation W43812505213 @default.
• W4381250521 hasOpenAccess W4381250521 @default.
• W4381250521 hasPrimaryLocation W43812505211 @default.
• W4381250521 hasRelatedWork W1593354505 @default.
• W4381250521 hasRelatedWork W1963986408 @default.
• W4381250521 hasRelatedWork W2061011797 @default.
• W4381250521 hasRelatedWork W2148787589 @default.
• W4381250521 hasRelatedWork W2243773866 @default.
• W4381250521 hasRelatedWork W2796481166 @default.
• W4381250521 hasRelatedWork W3046303201 @default.
• W4381250521 hasRelatedWork W66126747 @default.
• W4381250521 hasRelatedWork W73328466 @default.
• W4381250521 hasRelatedWork W2157723426 @default.
• W4381250521 hasVolume "11" @default.
• W4381250521 isParatext "false" @default.
• W4381250521 isRetracted "false" @default.
• W4381250521 workType "article" @default.