FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2009225465> ?p ?o ?g. }

• W2009225465 endingPage "399" @default.
• W2009225465 startingPage "373" @default.
• W2009225465 abstract "In the analysis and prediction of real-world systems, two of the key problems are nonstationarity (often in the form of switching between regimes), and overfitting (particularly serious for noisy processes). This article addresses these problems using gated experts, consisting of a (nonlinear) gating network, and several (also nonlinear) competing experts. Each expert learns to predict the conditional mean, and each expert adapts its width to match the noise level in its regime. The gating network learns to predict the probability of each expert, given the input. This article focuses on the case where the gating network bases its decision on information from the inputs. This can be contrasted to hidden Markov models where the decision is based on the previous state(s) (i.e. on the output of the gating network at the previous time step), as well as to averaging over several predictors. In contrast, gated experts soft-partition the input space, only learning to model their region. This article discusses the underlying statistical assumptions, derives the weight update rules, and compares the performance of gated experts to standard methods on three time series: (1) a computer-generated series, obtained by randomly switching between two nonlinear processes; (2) a time series from the Santa Fe Time Series Competition (the light intensity of a laser in chaotic state); and (3) the daily electricity demand of France, a real-world multivariate problem with structure on several time scales. The main results are: (1) the gating network correctly discovers the different regimes of the process; (2) the widths associated with each expert are important for the segmentation task (and they can be used to characterize the sub-processes); and (3) there is less overfitting compared to single networks (homogeneous multilayer perceptrons), since the experts learn to match their variances to the (local) noise levels. This can be viewed as matching the local complexity of the model to the local complexity of the data." @default.
• W2009225465 created "2016-06-24" @default.
• W2009225465 creator A5034697967 @default.
• W2009225465 creator A5037027451 @default.
• W2009225465 creator A5049627979 @default.
• W2009225465 date "1995-12-01" @default.
• W2009225465 modified "2023-10-02" @default.
• W2009225465 title "NONLINEAR GATED EXPERTS FOR TIME SERIES: DISCOVERING REGIMES AND AVOIDING OVERFITTING" @default.
• W2009225465 cites W1528905581 @default.
• W2009225465 cites W1573546415 @default.
• W2009225465 cites W1576326745 @default.
• W2009225465 cites W1606326889 @default.
• W2009225465 cites W1709719830 @default.
• W2009225465 cites W1944524590 @default.
• W2009225465 cites W1981367467 @default.
• W2009225465 cites W1995930498 @default.
• W2009225465 cites W1999642580 @default.
• W2009225465 cites W2006430387 @default.
• W2009225465 cites W2009588932 @default.
• W2009225465 cites W2011277999 @default.
• W2009225465 cites W2022852240 @default.
• W2009225465 cites W2025653905 @default.
• W2009225465 cites W2025887562 @default.
• W2009225465 cites W2049633694 @default.
• W2009225465 cites W2052441401 @default.
• W2009225465 cites W2095539364 @default.
• W2009225465 cites W2102201073 @default.
• W2009225465 cites W2106644119 @default.
• W2009225465 cites W2113021507 @default.
• W2009225465 cites W2125838338 @default.
• W2009225465 cites W2141899961 @default.
• W2009225465 cites W2143956139 @default.
• W2009225465 cites W2161523118 @default.
• W2009225465 cites W2263098707 @default.
• W2009225465 cites W2727300753 @default.
• W2009225465 cites W2797066837 @default.
• W2009225465 cites W2896916590 @default.
• W2009225465 cites W2981264952 @default.
• W2009225465 cites W3123887400 @default.
• W2009225465 cites W3125890304 @default.
• W2009225465 cites W3146166473 @default.
• W2009225465 cites W94523489 @default.
• W2009225465 doi "https://doi.org/10.1142/s0129065795000251" @default.
• W2009225465 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/8963468" @default.
• W2009225465 hasPublicationYear "1995" @default.
• W2009225465 type Work @default.
• W2009225465 sameAs 2009225465 @default.
• W2009225465 citedByCount "239" @default.
• W2009225465 countsByYear W20092254652012 @default.
• W2009225465 countsByYear W20092254652013 @default.
• W2009225465 countsByYear W20092254652014 @default.
• W2009225465 countsByYear W20092254652015 @default.
• W2009225465 countsByYear W20092254652016 @default.
• W2009225465 countsByYear W20092254652017 @default.
• W2009225465 countsByYear W20092254652018 @default.
• W2009225465 countsByYear W20092254652019 @default.
• W2009225465 countsByYear W20092254652020 @default.
• W2009225465 countsByYear W20092254652021 @default.
• W2009225465 countsByYear W20092254652022 @default.
• W2009225465 countsByYear W20092254652023 @default.
• W2009225465 crossrefType "journal-article" @default.
• W2009225465 hasAuthorship W2009225465A5034697967 @default.
• W2009225465 hasAuthorship W2009225465A5037027451 @default.
• W2009225465 hasAuthorship W2009225465A5049627979 @default.
• W2009225465 hasConcept C119857082 @default.
• W2009225465 hasConcept C121332964 @default.
• W2009225465 hasConcept C124101348 @default.
• W2009225465 hasConcept C143724316 @default.
• W2009225465 hasConcept C151406439 @default.
• W2009225465 hasConcept C151730666 @default.
• W2009225465 hasConcept C154945302 @default.
• W2009225465 hasConcept C158622935 @default.
• W2009225465 hasConcept C194544171 @default.
• W2009225465 hasConcept C22019652 @default.
• W2009225465 hasConcept C41008148 @default.
• W2009225465 hasConcept C42407357 @default.
• W2009225465 hasConcept C50644808 @default.
• W2009225465 hasConcept C62520636 @default.
• W2009225465 hasConcept C86803240 @default.
• W2009225465 hasConcept C98763669 @default.
• W2009225465 hasConceptScore W2009225465C119857082 @default.
• W2009225465 hasConceptScore W2009225465C121332964 @default.
• W2009225465 hasConceptScore W2009225465C124101348 @default.
• W2009225465 hasConceptScore W2009225465C143724316 @default.
• W2009225465 hasConceptScore W2009225465C151406439 @default.
• W2009225465 hasConceptScore W2009225465C151730666 @default.
• W2009225465 hasConceptScore W2009225465C154945302 @default.
• W2009225465 hasConceptScore W2009225465C158622935 @default.
• W2009225465 hasConceptScore W2009225465C194544171 @default.
• W2009225465 hasConceptScore W2009225465C22019652 @default.
• W2009225465 hasConceptScore W2009225465C41008148 @default.
• W2009225465 hasConceptScore W2009225465C42407357 @default.
• W2009225465 hasConceptScore W2009225465C50644808 @default.
• W2009225465 hasConceptScore W2009225465C62520636 @default.
• W2009225465 hasConceptScore W2009225465C86803240 @default.
• W2009225465 hasConceptScore W2009225465C98763669 @default.
• W2009225465 hasIssue "04" @default.
• W2009225465 hasLocation W20092254651 @default.

• next