Matches in SemOpenAlex for { <https://semopenalex.org/work/W3200217834> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W3200217834 abstract "Deep neural networks (DNNs) are capable of nonlinear learning and achieve high accuracy in many machine learning tasks. However, their performance could degrade due to noisy data, insufficient data amount, or data distributional shifts. In this work, we propose a novel analysis approach to interpret DNNs performance degradation by employing Bayesian uncertainty estimates. In essence, we derive a new probabilistic modeling method to separate and measure three major sources that cause performance degradation, namely data uncertainty, model uncertainty and distributional uncertainty. By performing sensitivity analysis on the decomposed uncertainty estimates, we can analyze which features influence each type of uncertainty. We validate and demonstrate the usefulness of our analysis on real-world datasets in both classification and regression tasks. Our method improves the interpretability of black-box DNNs." @default.
- W3200217834 created "2021-09-27" @default.
- W3200217834 creator A5001154049 @default.
- W3200217834 creator A5024991309 @default.
- W3200217834 date "2021-07-18" @default.
- W3200217834 modified "2023-10-16" @default.
- W3200217834 title "Model Performance Inspection of Deep Neural Networks by Decomposing Bayesian Uncertainty Estimates" @default.
- W3200217834 cites W2159778339 @default.
- W3200217834 cites W2600383743 @default.
- W3200217834 cites W2772430408 @default.
- W3200217834 cites W2884298516 @default.
- W3200217834 cites W2886614482 @default.
- W3200217834 cites W2963215553 @default.
- W3200217834 cites W2963677766 @default.
- W3200217834 cites W2964153729 @default.
- W3200217834 cites W2964317943 @default.
- W3200217834 cites W3007351208 @default.
- W3200217834 cites W3014339631 @default.
- W3200217834 cites W3093205559 @default.
- W3200217834 doi "https://doi.org/10.1109/ijcnn52387.2021.9533764" @default.
- W3200217834 hasPublicationYear "2021" @default.
- W3200217834 type Work @default.
- W3200217834 sameAs 3200217834 @default.
- W3200217834 citedByCount "0" @default.
- W3200217834 crossrefType "proceedings-article" @default.
- W3200217834 hasAuthorship W3200217834A5001154049 @default.
- W3200217834 hasAuthorship W3200217834A5024991309 @default.
- W3200217834 hasConcept C107673813 @default.
- W3200217834 hasConcept C119857082 @default.
- W3200217834 hasConcept C124101348 @default.
- W3200217834 hasConcept C127413603 @default.
- W3200217834 hasConcept C154945302 @default.
- W3200217834 hasConcept C177803969 @default.
- W3200217834 hasConcept C21200559 @default.
- W3200217834 hasConcept C24326235 @default.
- W3200217834 hasConcept C2780009758 @default.
- W3200217834 hasConcept C2781067378 @default.
- W3200217834 hasConcept C2984842247 @default.
- W3200217834 hasConcept C32230216 @default.
- W3200217834 hasConcept C33724603 @default.
- W3200217834 hasConcept C41008148 @default.
- W3200217834 hasConcept C44154836 @default.
- W3200217834 hasConcept C49937458 @default.
- W3200217834 hasConcept C50644808 @default.
- W3200217834 hasConcept C94966114 @default.
- W3200217834 hasConceptScore W3200217834C107673813 @default.
- W3200217834 hasConceptScore W3200217834C119857082 @default.
- W3200217834 hasConceptScore W3200217834C124101348 @default.
- W3200217834 hasConceptScore W3200217834C127413603 @default.
- W3200217834 hasConceptScore W3200217834C154945302 @default.
- W3200217834 hasConceptScore W3200217834C177803969 @default.
- W3200217834 hasConceptScore W3200217834C21200559 @default.
- W3200217834 hasConceptScore W3200217834C24326235 @default.
- W3200217834 hasConceptScore W3200217834C2780009758 @default.
- W3200217834 hasConceptScore W3200217834C2781067378 @default.
- W3200217834 hasConceptScore W3200217834C2984842247 @default.
- W3200217834 hasConceptScore W3200217834C32230216 @default.
- W3200217834 hasConceptScore W3200217834C33724603 @default.
- W3200217834 hasConceptScore W3200217834C41008148 @default.
- W3200217834 hasConceptScore W3200217834C44154836 @default.
- W3200217834 hasConceptScore W3200217834C49937458 @default.
- W3200217834 hasConceptScore W3200217834C50644808 @default.
- W3200217834 hasConceptScore W3200217834C94966114 @default.
- W3200217834 hasLocation W32002178341 @default.
- W3200217834 hasOpenAccess W3200217834 @default.
- W3200217834 hasPrimaryLocation W32002178341 @default.
- W3200217834 hasRelatedWork W1807198 @default.
- W3200217834 hasRelatedWork W2701911 @default.
- W3200217834 hasRelatedWork W3891032 @default.
- W3200217834 hasRelatedWork W4179021 @default.
- W3200217834 hasRelatedWork W5005022 @default.
- W3200217834 hasRelatedWork W5508898 @default.
- W3200217834 hasRelatedWork W758187 @default.
- W3200217834 hasRelatedWork W7965686 @default.
- W3200217834 hasRelatedWork W8021486 @default.
- W3200217834 hasRelatedWork W8978941 @default.
- W3200217834 isParatext "false" @default.
- W3200217834 isRetracted "false" @default.
- W3200217834 magId "3200217834" @default.
- W3200217834 workType "article" @default.