FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2948194985> ?p ?o ?g. }

• W2948194985 abstract "Modern machine learning methods including deep learning have achieved great success in predictive accuracy for supervised learning tasks, but may still fall short in giving useful estimates of their predictive {em uncertainty}. Quantifying uncertainty is especially critical in real-world settings, which often involve input distributions that are shifted from the training distribution due to a variety of factors including sample bias and non-stationarity. In such settings, well calibrated uncertainty estimates convey information about when a model's output should (or should not) be trusted. Many probabilistic deep learning methods, including Bayesian-and non-Bayesian methods, have been proposed in the literature for quantifying predictive uncertainty, but to our knowledge there has not previously been a rigorous large-scale empirical comparison of these methods under dataset shift. We present a large-scale benchmark of existing state-of-the-art methods on classification problems and investigate the effect of dataset shift on accuracy and calibration. We find that traditional post-hoc calibration does indeed fall short, as do several other previous methods. However, some methods that marginalize over models give surprisingly strong results across a broad spectrum of tasks." @default.
• W2948194985 created "2019-06-14" @default.
• W2948194985 creator A5000291599 @default.
• W2948194985 creator A5012859570 @default.
• W2948194985 creator A5016611202 @default.
• W2948194985 creator A5021577029 @default.
• W2948194985 creator A5026399928 @default.
• W2948194985 creator A5044982268 @default.
• W2948194985 creator A5045421827 @default.
• W2948194985 creator A5055928795 @default.
• W2948194985 creator A5059296721 @default.
• W2948194985 date "2019-06-06" @default.
• W2948194985 modified "2023-10-05" @default.
• W2948194985 title "Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift" @default.
• W2948194985 cites W1493526108 @default.
• W2948194985 cites W1533660737 @default.
• W2948194985 cites W1599263113 @default.
• W2948194985 cites W1618905105 @default.
• W2948194985 cites W1719489212 @default.
• W2948194985 cites W1826234144 @default.
• W2948194985 cites W2025720061 @default.
• W2948194985 cites W2064675550 @default.
• W2948194985 cites W2073241381 @default.
• W2948194985 cites W2108501770 @default.
• W2948194985 cites W2108598243 @default.
• W2948194985 cites W2108677974 @default.
• W2948194985 cites W2112796928 @default.
• W2948194985 cites W2117670920 @default.
• W2948194985 cites W2123098109 @default.
• W2948194985 cites W2158471180 @default.
• W2948194985 cites W2162651021 @default.
• W2948194985 cites W2164411961 @default.
• W2948194985 cites W2167433878 @default.
• W2948194985 cites W2194775991 @default.
• W2948194985 cites W2254249950 @default.
• W2948194985 cites W2335728318 @default.
• W2948194985 cites W2342840547 @default.
• W2948194985 cites W2462906003 @default.
• W2948194985 cites W2531327146 @default.
• W2948194985 cites W2592505114 @default.
• W2948194985 cites W2600383743 @default.
• W2948194985 cites W2624413595 @default.
• W2948194985 cites W2732547613 @default.
• W2948194985 cites W2785606994 @default.
• W2948194985 cites W2810676804 @default.
• W2948194985 cites W2867167548 @default.
• W2948194985 cites W2881747041 @default.
• W2948194985 cites W2892375480 @default.
• W2948194985 cites W2898631838 @default.
• W2948194985 cites W2903995489 @default.
• W2948194985 cites W2907020378 @default.
• W2948194985 cites W2912399224 @default.
• W2948194985 cites W2949416428 @default.
• W2948194985 cites W2949738940 @default.
• W2948194985 cites W2950621961 @default.
• W2948194985 cites W2951266961 @default.
• W2948194985 cites W2951714314 @default.
• W2948194985 cites W2952300048 @default.
• W2948194985 cites W2963060032 @default.
• W2948194985 cites W2963238274 @default.
• W2948194985 cites W2963613748 @default.
• W2948194985 cites W2963693742 @default.
• W2948194985 cites W2963938771 @default.
• W2948194985 cites W2964059111 @default.
• W2948194985 cites W2964121744 @default.
• W2948194985 cites W2964212410 @default.
• W2948194985 cites W2964330179 @default.
• W2948194985 cites W2969609970 @default.
• W2948194985 cites W2970946347 @default.
• W2948194985 cites W3118608800 @default.
• W2948194985 hasPublicationYear "2019" @default.
• W2948194985 type Work @default.
• W2948194985 sameAs 2948194985 @default.
• W2948194985 citedByCount "67" @default.
• W2948194985 countsByYear W29481949852019 @default.
• W2948194985 countsByYear W29481949852020 @default.
• W2948194985 countsByYear W29481949852021 @default.
• W2948194985 crossrefType "posted-content" @default.
• W2948194985 hasAuthorship W2948194985A5000291599 @default.
• W2948194985 hasAuthorship W2948194985A5012859570 @default.
• W2948194985 hasAuthorship W2948194985A5016611202 @default.
• W2948194985 hasAuthorship W2948194985A5021577029 @default.
• W2948194985 hasAuthorship W2948194985A5026399928 @default.
• W2948194985 hasAuthorship W2948194985A5044982268 @default.
• W2948194985 hasAuthorship W2948194985A5045421827 @default.
• W2948194985 hasAuthorship W2948194985A5055928795 @default.
• W2948194985 hasAuthorship W2948194985A5059296721 @default.
• W2948194985 hasConcept C105795698 @default.
• W2948194985 hasConcept C107673813 @default.
• W2948194985 hasConcept C108583219 @default.
• W2948194985 hasConcept C111472728 @default.
• W2948194985 hasConcept C119857082 @default.
• W2948194985 hasConcept C121332964 @default.
• W2948194985 hasConcept C124101348 @default.
• W2948194985 hasConcept C13280743 @default.
• W2948194985 hasConcept C138885662 @default.
• W2948194985 hasConcept C149782125 @default.
• W2948194985 hasConcept C154945302 @default.
• W2948194985 hasConcept C165838908 @default.
• W2948194985 hasConcept C185798385 @default.

• next