Matches in SemOpenAlex for { <https://semopenalex.org/work/W3016296568> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W3016296568 abstract "Setting hyperparameters is a complex and hard task whentraining a neural network. Neural networks performingbadly can be the effect of a sub-optimal hyperparameterssetup before training. This study aims to examine theeffects of hyperparameters on the training of a residualnetwork, while using a small dataset. This network is aSE-ResNet-50 that is being used to recognize emotionsfrom facial expressions. Different methods, both constantvalues as well as schedules, are compared based on thevalidation loss and validation accuracy.For interesting outcomes, an examination is done to seewhether these methods transfer well to a similar datasetin the same context of emotion recognition in facial ex-pressions.The study shows that low batch sizes contribute to a goodperformance, but tend to result in an unstable network. Abalance must be found between a high accuracy and a sta-ble training. Learning rate schedules that use small stepsizes outperform those which make larger adjustments.Momentum is a valuable addition, but too high of a mo-mentum shows signs of overfitting in the loss graph. Acyclical momentum causes the training to become unsta-ble, this can be reduced by using higher batch sizes." @default.
- W3016296568 created "2020-04-24" @default.
- W3016296568 creator A5077494237 @default.
- W3016296568 date "2020-01-01" @default.
- W3016296568 modified "2023-09-27" @default.
- W3016296568 title "Examining the effect of hyperparameters on the trainingof a residual network for emotion recognition" @default.
- W3016296568 hasPublicationYear "2020" @default.
- W3016296568 type Work @default.
- W3016296568 sameAs 3016296568 @default.
- W3016296568 citedByCount "0" @default.
- W3016296568 crossrefType "dissertation" @default.
- W3016296568 hasAuthorship W3016296568A5077494237 @default.
- W3016296568 hasConcept C11413529 @default.
- W3016296568 hasConcept C119857082 @default.
- W3016296568 hasConcept C127413603 @default.
- W3016296568 hasConcept C150899416 @default.
- W3016296568 hasConcept C151730666 @default.
- W3016296568 hasConcept C153180895 @default.
- W3016296568 hasConcept C154945302 @default.
- W3016296568 hasConcept C155512373 @default.
- W3016296568 hasConcept C201995342 @default.
- W3016296568 hasConcept C22019652 @default.
- W3016296568 hasConcept C2779343474 @default.
- W3016296568 hasConcept C2780451532 @default.
- W3016296568 hasConcept C41008148 @default.
- W3016296568 hasConcept C50644808 @default.
- W3016296568 hasConcept C8642999 @default.
- W3016296568 hasConcept C86803240 @default.
- W3016296568 hasConceptScore W3016296568C11413529 @default.
- W3016296568 hasConceptScore W3016296568C119857082 @default.
- W3016296568 hasConceptScore W3016296568C127413603 @default.
- W3016296568 hasConceptScore W3016296568C150899416 @default.
- W3016296568 hasConceptScore W3016296568C151730666 @default.
- W3016296568 hasConceptScore W3016296568C153180895 @default.
- W3016296568 hasConceptScore W3016296568C154945302 @default.
- W3016296568 hasConceptScore W3016296568C155512373 @default.
- W3016296568 hasConceptScore W3016296568C201995342 @default.
- W3016296568 hasConceptScore W3016296568C22019652 @default.
- W3016296568 hasConceptScore W3016296568C2779343474 @default.
- W3016296568 hasConceptScore W3016296568C2780451532 @default.
- W3016296568 hasConceptScore W3016296568C41008148 @default.
- W3016296568 hasConceptScore W3016296568C50644808 @default.
- W3016296568 hasConceptScore W3016296568C8642999 @default.
- W3016296568 hasConceptScore W3016296568C86803240 @default.
- W3016296568 hasLocation W30162965681 @default.
- W3016296568 hasOpenAccess W3016296568 @default.
- W3016296568 hasPrimaryLocation W30162965681 @default.
- W3016296568 hasRelatedWork W126232561 @default.
- W3016296568 hasRelatedWork W1545175472 @default.
- W3016296568 hasRelatedWork W1596504373 @default.
- W3016296568 hasRelatedWork W2736081685 @default.
- W3016296568 hasRelatedWork W2752295241 @default.
- W3016296568 hasRelatedWork W2797402435 @default.
- W3016296568 hasRelatedWork W2912626019 @default.
- W3016296568 hasRelatedWork W2979906247 @default.
- W3016296568 hasRelatedWork W2995472160 @default.
- W3016296568 hasRelatedWork W3010974259 @default.
- W3016296568 hasRelatedWork W3041928889 @default.
- W3016296568 hasRelatedWork W3088551846 @default.
- W3016296568 hasRelatedWork W3127385465 @default.
- W3016296568 hasRelatedWork W3130848728 @default.
- W3016296568 hasRelatedWork W3161021613 @default.
- W3016296568 hasRelatedWork W3165499928 @default.
- W3016296568 hasRelatedWork W3183912051 @default.
- W3016296568 hasRelatedWork W3201809877 @default.
- W3016296568 hasRelatedWork W3205994208 @default.
- W3016296568 hasRelatedWork W996801710 @default.
- W3016296568 isParatext "false" @default.
- W3016296568 isRetracted "false" @default.
- W3016296568 magId "3016296568" @default.
- W3016296568 workType "dissertation" @default.