Matches in SemOpenAlex for { <https://semopenalex.org/work/W3090505727> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W3090505727 endingPage "833" @default.
- W3090505727 startingPage "824" @default.
- W3090505727 abstract "Deep learning-based algorithms have shown great promise for assisting pathologists in detecting lymph node metastases when evaluated based on their predictive accuracy. However, for clinical adoption, we need to know what happens when the test set dramatically changes from the training distribution. In such settings, we should estimate the uncertainty of the predictions, so we know when to trust the model (and when not to). Here, we i) investigate current popular methods for improving the calibration of predictive uncertainty, and ii) compare the performance and calibration of the methods under clinically relevant in-distribution dataset shifts. Furthermore, we iii) evaluate their performance on the task of out-of-distribution detection of a different histological cancer type not seen during training. Of the investigated methods, we show that deep ensembles are more robust in respect of both performance and calibration for in-distribution dataset shifts and allows us to better detect incorrect predictions. Our results also demonstrate that current methods for uncertainty quantification are not necessarily able to detect all dataset shifts, and we emphasize the importance of monitoring and controlling the input distribution when deploying deep learning for digital pathology." @default.
- W3090505727 created "2020-10-08" @default.
- W3090505727 creator A5001215398 @default.
- W3090505727 creator A5005339159 @default.
- W3090505727 creator A5009813954 @default.
- W3090505727 creator A5031953389 @default.
- W3090505727 creator A5072186051 @default.
- W3090505727 creator A5088556639 @default.
- W3090505727 date "2020-01-01" @default.
- W3090505727 modified "2023-10-06" @default.
- W3090505727 title "Can You Trust Predictive Uncertainty Under Real Dataset Shifts in Digital Pathology?" @default.
- W3090505727 cites W2194775991 @default.
- W3090505727 cites W2594258169 @default.
- W3090505727 cites W2772723798 @default.
- W3090505727 cites W2805886241 @default.
- W3090505727 cites W2889232360 @default.
- W3090505727 cites W2894917609 @default.
- W3090505727 cites W2897434820 @default.
- W3090505727 cites W2969278648 @default.
- W3090505727 cites W2970121940 @default.
- W3090505727 doi "https://doi.org/10.1007/978-3-030-59710-8_80" @default.
- W3090505727 hasPublicationYear "2020" @default.
- W3090505727 type Work @default.
- W3090505727 sameAs 3090505727 @default.
- W3090505727 citedByCount "16" @default.
- W3090505727 countsByYear W30905057272020 @default.
- W3090505727 countsByYear W30905057272021 @default.
- W3090505727 countsByYear W30905057272022 @default.
- W3090505727 countsByYear W30905057272023 @default.
- W3090505727 crossrefType "book-chapter" @default.
- W3090505727 hasAuthorship W3090505727A5001215398 @default.
- W3090505727 hasAuthorship W3090505727A5005339159 @default.
- W3090505727 hasAuthorship W3090505727A5009813954 @default.
- W3090505727 hasAuthorship W3090505727A5031953389 @default.
- W3090505727 hasAuthorship W3090505727A5072186051 @default.
- W3090505727 hasAuthorship W3090505727A5088556639 @default.
- W3090505727 hasBestOaLocation W30905057272 @default.
- W3090505727 hasConcept C105795698 @default.
- W3090505727 hasConcept C108583219 @default.
- W3090505727 hasConcept C119857082 @default.
- W3090505727 hasConcept C124101348 @default.
- W3090505727 hasConcept C127413603 @default.
- W3090505727 hasConcept C154945302 @default.
- W3090505727 hasConcept C162324750 @default.
- W3090505727 hasConcept C165838908 @default.
- W3090505727 hasConcept C169903167 @default.
- W3090505727 hasConcept C177264268 @default.
- W3090505727 hasConcept C187736073 @default.
- W3090505727 hasConcept C199360897 @default.
- W3090505727 hasConcept C2777522853 @default.
- W3090505727 hasConcept C2780451532 @default.
- W3090505727 hasConcept C33923547 @default.
- W3090505727 hasConcept C41008148 @default.
- W3090505727 hasConcept C51632099 @default.
- W3090505727 hasConcept C62611344 @default.
- W3090505727 hasConcept C66938386 @default.
- W3090505727 hasConceptScore W3090505727C105795698 @default.
- W3090505727 hasConceptScore W3090505727C108583219 @default.
- W3090505727 hasConceptScore W3090505727C119857082 @default.
- W3090505727 hasConceptScore W3090505727C124101348 @default.
- W3090505727 hasConceptScore W3090505727C127413603 @default.
- W3090505727 hasConceptScore W3090505727C154945302 @default.
- W3090505727 hasConceptScore W3090505727C162324750 @default.
- W3090505727 hasConceptScore W3090505727C165838908 @default.
- W3090505727 hasConceptScore W3090505727C169903167 @default.
- W3090505727 hasConceptScore W3090505727C177264268 @default.
- W3090505727 hasConceptScore W3090505727C187736073 @default.
- W3090505727 hasConceptScore W3090505727C199360897 @default.
- W3090505727 hasConceptScore W3090505727C2777522853 @default.
- W3090505727 hasConceptScore W3090505727C2780451532 @default.
- W3090505727 hasConceptScore W3090505727C33923547 @default.
- W3090505727 hasConceptScore W3090505727C41008148 @default.
- W3090505727 hasConceptScore W3090505727C51632099 @default.
- W3090505727 hasConceptScore W3090505727C62611344 @default.
- W3090505727 hasConceptScore W3090505727C66938386 @default.
- W3090505727 hasLocation W30905057271 @default.
- W3090505727 hasLocation W30905057272 @default.
- W3090505727 hasOpenAccess W3090505727 @default.
- W3090505727 hasPrimaryLocation W30905057271 @default.
- W3090505727 hasRelatedWork W2792951589 @default.
- W3090505727 hasRelatedWork W3099765033 @default.
- W3090505727 hasRelatedWork W3150234497 @default.
- W3090505727 hasRelatedWork W3165388794 @default.
- W3090505727 hasRelatedWork W3201070945 @default.
- W3090505727 hasRelatedWork W4223943233 @default.
- W3090505727 hasRelatedWork W4287181775 @default.
- W3090505727 hasRelatedWork W4312200629 @default.
- W3090505727 hasRelatedWork W4360585206 @default.
- W3090505727 hasRelatedWork W4380075502 @default.
- W3090505727 isParatext "false" @default.
- W3090505727 isRetracted "false" @default.
- W3090505727 magId "3090505727" @default.
- W3090505727 workType "book-chapter" @default.