Matches in SemOpenAlex for { <https://semopenalex.org/work/W4211173885> ?p ?o ?g. }
Showing items 1 to 56 of
56
with 100 items per page.
- W4211173885 abstract "With the development of deep neural networks, in order to deal with overfitting, many researchers have proposed various methods. Dropout is a regularization technique and is often used in deep neural networks, especially in the field of image classification, so that the classification accuracy can be improved. However, these methods rarely consider that the change of parameters can affect the role of neurons in neural networks. In this paper, we propose a dynamic dropout to use a parameter gradient-based method, which utilizes the cumulative sum of the absolute value of the gradient of each hidden unit corresponding to all parameters. Specifically, each neuron has a corresponding gradient under the circumstances, and then we only do dropout on neurons which have large gradient value. The purpose is to let neurons with small gradient value learn more information. We evaluated the effectiveness of Dynamic Dropout experiments on multiple datasets, including MNIST, CIFAR-10, and CIFAR-100, the models are convolutional neural networks." @default.
- W4211173885 created "2022-02-13" @default.
- W4211173885 creator A5002495815 @default.
- W4211173885 creator A5033732336 @default.
- W4211173885 creator A5064414647 @default.
- W4211173885 creator A5079873685 @default.
- W4211173885 date "2021-11-01" @default.
- W4211173885 modified "2023-09-29" @default.
- W4211173885 title "Gradient-based Dynamic Dropout" @default.
- W4211173885 doi "https://doi.org/10.1109/cis54983.2021.00037" @default.
- W4211173885 hasPublicationYear "2021" @default.
- W4211173885 type Work @default.
- W4211173885 citedByCount "0" @default.
- W4211173885 crossrefType "proceedings-article" @default.
- W4211173885 hasAuthorship W4211173885A5002495815 @default.
- W4211173885 hasAuthorship W4211173885A5033732336 @default.
- W4211173885 hasAuthorship W4211173885A5064414647 @default.
- W4211173885 hasAuthorship W4211173885A5079873685 @default.
- W4211173885 hasConcept C108583219 @default.
- W4211173885 hasConcept C119857082 @default.
- W4211173885 hasConcept C153180895 @default.
- W4211173885 hasConcept C154945302 @default.
- W4211173885 hasConcept C190502265 @default.
- W4211173885 hasConcept C22019652 @default.
- W4211173885 hasConcept C2776135515 @default.
- W4211173885 hasConcept C2776145597 @default.
- W4211173885 hasConcept C41008148 @default.
- W4211173885 hasConcept C50644808 @default.
- W4211173885 hasConcept C81363708 @default.
- W4211173885 hasConceptScore W4211173885C108583219 @default.
- W4211173885 hasConceptScore W4211173885C119857082 @default.
- W4211173885 hasConceptScore W4211173885C153180895 @default.
- W4211173885 hasConceptScore W4211173885C154945302 @default.
- W4211173885 hasConceptScore W4211173885C190502265 @default.
- W4211173885 hasConceptScore W4211173885C22019652 @default.
- W4211173885 hasConceptScore W4211173885C2776135515 @default.
- W4211173885 hasConceptScore W4211173885C2776145597 @default.
- W4211173885 hasConceptScore W4211173885C41008148 @default.
- W4211173885 hasConceptScore W4211173885C50644808 @default.
- W4211173885 hasConceptScore W4211173885C81363708 @default.
- W4211173885 hasLocation W42111738851 @default.
- W4211173885 hasOpenAccess W4211173885 @default.
- W4211173885 hasPrimaryLocation W42111738851 @default.
- W4211173885 hasRelatedWork W2901800056 @default.
- W4211173885 hasRelatedWork W3099765033 @default.
- W4211173885 hasRelatedWork W3116689448 @default.
- W4211173885 hasRelatedWork W3128220493 @default.
- W4211173885 hasRelatedWork W3156786002 @default.
- W4211173885 hasRelatedWork W3186840088 @default.
- W4211173885 hasRelatedWork W3186919929 @default.
- W4211173885 hasRelatedWork W4287064118 @default.
- W4211173885 hasRelatedWork W4309224979 @default.
- W4211173885 hasRelatedWork W4317374280 @default.
- W4211173885 isParatext "false" @default.
- W4211173885 isRetracted "false" @default.
- W4211173885 workType "article" @default.