Matches in SemOpenAlex for { <https://semopenalex.org/work/W3201739046> ?p ?o ?g. }
- W3201739046 abstract "Uncertainty quantification of machine learning and deep learning methods plays an important role in enhancing trust to the obtained result. In recent years, a numerous number of uncertainty quantification methods have been introduced. Monte Carlo dropout (MC-Dropout) is one of the most well-known techniques to quantify uncertainty in deep learning methods. In this study, we propose two new loss functions by combining cross entropy with Expected Calibration Error (ECE) and Predictive Entropy (PE). The obtained results clearly show that the new proposed loss functions lead to having a calibrated MC-Dropout method. Our results confirmed the great impact of the new hybrid loss functions for minimising the overlap between the distributions of uncertainty estimates for correct and incorrect predictions without sacrificing the model's overall performance." @default.
- W3201739046 created "2021-10-11" @default.
- W3201739046 creator A5014000715 @default.
- W3201739046 creator A5015293969 @default.
- W3201739046 creator A5029751075 @default.
- W3201739046 creator A5036354283 @default.
- W3201739046 creator A5045661979 @default.
- W3201739046 creator A5059557438 @default.
- W3201739046 creator A5072780166 @default.
- W3201739046 date "2021-10-07" @default.
- W3201739046 modified "2023-09-23" @default.
- W3201739046 title "Improving MC-Dropout Uncertainty Estimates with Calibration Error-based Optimization." @default.
- W3201739046 cites W2166851633 @default.
- W3201739046 cites W2204383650 @default.
- W3201739046 cites W2560321925 @default.
- W3201739046 cites W2610571781 @default.
- W3201739046 cites W2626967530 @default.
- W3201739046 cites W2679674811 @default.
- W3201739046 cites W2807414627 @default.
- W3201739046 cites W2901493566 @default.
- W3201739046 cites W2907425548 @default.
- W3201739046 cites W2922164623 @default.
- W3201739046 cites W2950469931 @default.
- W3201739046 cites W2950517871 @default.
- W3201739046 cites W2963842958 @default.
- W3201739046 cites W2963977107 @default.
- W3201739046 cites W2964059111 @default.
- W3201739046 cites W2972188071 @default.
- W3201739046 cites W2978575375 @default.
- W3201739046 cites W3006617515 @default.
- W3201739046 cites W3012312660 @default.
- W3201739046 cites W3031806488 @default.
- W3201739046 cites W3032297297 @default.
- W3201739046 cites W3034441884 @default.
- W3201739046 cites W3035136782 @default.
- W3201739046 cites W3037349563 @default.
- W3201739046 cites W3038042546 @default.
- W3201739046 cites W3044724994 @default.
- W3201739046 cites W3080764565 @default.
- W3201739046 cites W3114015798 @default.
- W3201739046 cites W3116063148 @default.
- W3201739046 cites W3118929067 @default.
- W3201739046 cites W3147476671 @default.
- W3201739046 cites W3158436118 @default.
- W3201739046 cites W3161020808 @default.
- W3201739046 cites W3161619868 @default.
- W3201739046 cites W3164178813 @default.
- W3201739046 cites W3165396901 @default.
- W3201739046 cites W3166008066 @default.
- W3201739046 cites W3170765241 @default.
- W3201739046 cites W3173854166 @default.
- W3201739046 cites W3174011424 @default.
- W3201739046 cites W3174145583 @default.
- W3201739046 cites W3175065557 @default.
- W3201739046 cites W3181597434 @default.
- W3201739046 cites W3184219842 @default.
- W3201739046 cites W3186353005 @default.
- W3201739046 cites W41222772 @default.
- W3201739046 hasPublicationYear "2021" @default.
- W3201739046 type Work @default.
- W3201739046 sameAs 3201739046 @default.
- W3201739046 citedByCount "0" @default.
- W3201739046 crossrefType "posted-content" @default.
- W3201739046 hasAuthorship W3201739046A5014000715 @default.
- W3201739046 hasAuthorship W3201739046A5015293969 @default.
- W3201739046 hasAuthorship W3201739046A5029751075 @default.
- W3201739046 hasAuthorship W3201739046A5036354283 @default.
- W3201739046 hasAuthorship W3201739046A5045661979 @default.
- W3201739046 hasAuthorship W3201739046A5059557438 @default.
- W3201739046 hasAuthorship W3201739046A5072780166 @default.
- W3201739046 hasConcept C105795698 @default.
- W3201739046 hasConcept C106301342 @default.
- W3201739046 hasConcept C119857082 @default.
- W3201739046 hasConcept C121332964 @default.
- W3201739046 hasConcept C137209882 @default.
- W3201739046 hasConcept C154945302 @default.
- W3201739046 hasConcept C165838908 @default.
- W3201739046 hasConcept C167981619 @default.
- W3201739046 hasConcept C19499675 @default.
- W3201739046 hasConcept C2776145597 @default.
- W3201739046 hasConcept C32230216 @default.
- W3201739046 hasConcept C33923547 @default.
- W3201739046 hasConcept C41008148 @default.
- W3201739046 hasConcept C62520636 @default.
- W3201739046 hasConcept C9679016 @default.
- W3201739046 hasConceptScore W3201739046C105795698 @default.
- W3201739046 hasConceptScore W3201739046C106301342 @default.
- W3201739046 hasConceptScore W3201739046C119857082 @default.
- W3201739046 hasConceptScore W3201739046C121332964 @default.
- W3201739046 hasConceptScore W3201739046C137209882 @default.
- W3201739046 hasConceptScore W3201739046C154945302 @default.
- W3201739046 hasConceptScore W3201739046C165838908 @default.
- W3201739046 hasConceptScore W3201739046C167981619 @default.
- W3201739046 hasConceptScore W3201739046C19499675 @default.
- W3201739046 hasConceptScore W3201739046C2776145597 @default.
- W3201739046 hasConceptScore W3201739046C32230216 @default.
- W3201739046 hasConceptScore W3201739046C33923547 @default.
- W3201739046 hasConceptScore W3201739046C41008148 @default.
- W3201739046 hasConceptScore W3201739046C62520636 @default.
- W3201739046 hasConceptScore W3201739046C9679016 @default.