FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2992766621> ?p ?o ?g. }

• W2992766621 endingPage "261" @default.
• W2992766621 startingPage "253" @default.
• W2992766621 abstract "Dropout, including Bernoulli dropout (equivalently random node fault) and multiplicative Gaussian noise (MGN) dropout (equivalently multiplicative node noise), has been a technique in training a neural network (NN) to achieve better performance. While simulation results have demonstrated its success, not many work has been done to explain why it works (or why it does not work). In this paper, the objective functions $$mathcal{L}({mathbf w})$$ of the learning algorithms with Bernoulli dropout and MGN dropout are derived and thus their regularization effects are analyzed. It is found that learning with Bernoulli dropout cannot improve the generalization of a NN if its weights are not scaled down after training. If we further let $$mathcal{J}({mathbf w})$$ be the desired measure of a NN with such inherent dropout, we clarify a misconception that $$mathcal{L}({mathbf w}) = mathcal{J}({mathbf w})$$. The model attained by learning with dropout is not the desired model that can tolerate such inherent dropout." @default.
• W2992766621 created "2019-12-13" @default.
• W2992766621 creator A5043836511 @default.
• W2992766621 creator A5053893418 @default.
• W2992766621 date "2019-01-01" @default.
• W2992766621 modified "2023-10-18" @default.
• W2992766621 title "Analysis on Dropout Regularization" @default.
• W2992766621 cites W1502369425 @default.
• W2992766621 cites W1964175594 @default.
• W2992766621 cites W1964929969 @default.
• W2992766621 cites W2003464224 @default.
• W2992766621 cites W2053006121 @default.
• W2992766621 cites W2057853719 @default.
• W2992766621 cites W2074623298 @default.
• W2992766621 cites W2079606844 @default.
• W2992766621 cites W2084163528 @default.
• W2992766621 cites W2111406701 @default.
• W2992766621 cites W2112882360 @default.
• W2992766621 cites W2123422798 @default.
• W2992766621 cites W2124136621 @default.
• W2992766621 cites W2145249131 @default.
• W2992766621 cites W2156084351 @default.
• W2992766621 cites W2160687178 @default.
• W2992766621 cites W2164370980 @default.
• W2992766621 cites W2171843308 @default.
• W2992766621 cites W2910723281 @default.
• W2992766621 cites W2965985861 @default.
• W2992766621 cites W2992766621 @default.
• W2992766621 doi "https://doi.org/10.1007/978-3-030-36802-9_28" @default.
• W2992766621 hasPublicationYear "2019" @default.
• W2992766621 type Work @default.
• W2992766621 sameAs 2992766621 @default.
• W2992766621 citedByCount "3" @default.
• W2992766621 countsByYear W29927666212019 @default.
• W2992766621 countsByYear W29927666212022 @default.
• W2992766621 countsByYear W29927666212023 @default.
• W2992766621 crossrefType "book-chapter" @default.
• W2992766621 hasAuthorship W2992766621A5043836511 @default.
• W2992766621 hasAuthorship W2992766621A5053893418 @default.
• W2992766621 hasConcept C105795698 @default.
• W2992766621 hasConcept C118615104 @default.
• W2992766621 hasConcept C119857082 @default.
• W2992766621 hasConcept C121332964 @default.
• W2992766621 hasConcept C122123141 @default.
• W2992766621 hasConcept C131021393 @default.
• W2992766621 hasConcept C13412647 @default.
• W2992766621 hasConcept C134306372 @default.
• W2992766621 hasConcept C152361515 @default.
• W2992766621 hasConcept C154945302 @default.
• W2992766621 hasConcept C163716315 @default.
• W2992766621 hasConcept C18015164 @default.
• W2992766621 hasConcept C185429906 @default.
• W2992766621 hasConcept C2776135515 @default.
• W2992766621 hasConcept C2776145597 @default.
• W2992766621 hasConcept C27956954 @default.
• W2992766621 hasConcept C28826006 @default.
• W2992766621 hasConcept C33923547 @default.
• W2992766621 hasConcept C41008148 @default.
• W2992766621 hasConcept C42747912 @default.
• W2992766621 hasConcept C62520636 @default.
• W2992766621 hasConcept C761482 @default.
• W2992766621 hasConcept C76155785 @default.
• W2992766621 hasConcept C97355855 @default.
• W2992766621 hasConceptScore W2992766621C105795698 @default.
• W2992766621 hasConceptScore W2992766621C118615104 @default.
• W2992766621 hasConceptScore W2992766621C119857082 @default.
• W2992766621 hasConceptScore W2992766621C121332964 @default.
• W2992766621 hasConceptScore W2992766621C122123141 @default.
• W2992766621 hasConceptScore W2992766621C131021393 @default.
• W2992766621 hasConceptScore W2992766621C13412647 @default.
• W2992766621 hasConceptScore W2992766621C134306372 @default.
• W2992766621 hasConceptScore W2992766621C152361515 @default.
• W2992766621 hasConceptScore W2992766621C154945302 @default.
• W2992766621 hasConceptScore W2992766621C163716315 @default.
• W2992766621 hasConceptScore W2992766621C18015164 @default.
• W2992766621 hasConceptScore W2992766621C185429906 @default.
• W2992766621 hasConceptScore W2992766621C2776135515 @default.
• W2992766621 hasConceptScore W2992766621C2776145597 @default.
• W2992766621 hasConceptScore W2992766621C27956954 @default.
• W2992766621 hasConceptScore W2992766621C28826006 @default.
• W2992766621 hasConceptScore W2992766621C33923547 @default.
• W2992766621 hasConceptScore W2992766621C41008148 @default.
• W2992766621 hasConceptScore W2992766621C42747912 @default.
• W2992766621 hasConceptScore W2992766621C62520636 @default.
• W2992766621 hasConceptScore W2992766621C761482 @default.
• W2992766621 hasConceptScore W2992766621C76155785 @default.
• W2992766621 hasConceptScore W2992766621C97355855 @default.
• W2992766621 hasLocation W29927666211 @default.
• W2992766621 hasOpenAccess W2992766621 @default.
• W2992766621 hasPrimaryLocation W29927666211 @default.
• W2992766621 hasRelatedWork W1499022509 @default.
• W2992766621 hasRelatedWork W1977150245 @default.
• W2992766621 hasRelatedWork W1982432439 @default.
• W2992766621 hasRelatedWork W2011124634 @default.
• W2992766621 hasRelatedWork W2088629913 @default.
• W2992766621 hasRelatedWork W2126687971 @default.
• W2992766621 hasRelatedWork W2155835781 @default.
• W2992766621 hasRelatedWork W2345434957 @default.

• next