Matches in SemOpenAlex for { <https://semopenalex.org/work/W3107312662> ?p ?o ?g. }
- W3107312662 endingPage "406" @default.
- W3107312662 startingPage "391" @default.
- W3107312662 abstract "The black-box nature of deep neural networks (DNNs) makes it impossible to understand why a particular output is produced, creating demand for “Explainable AI”. In this paper, we show that statistical fault localization (SFL) techniques from software engineering deliver high quality explanations of the outputs of DNNs, where we define an explanation as a minimal subset of features sufficient for making the same decision as for the original input. We present an algorithm and a tool called DeepCover, which synthesizes a ranking of the features of the inputs using SFL and constructs explanations for the decisions of the DNN based on this ranking. We compare explanations produced by DeepCover with those of the state-of-the-art tools gradcam, lime, shap, rise and extremal and show that explanations generated by DeepCover are consistently better across a broad set of experiments. On a benchmark set with known ground truth, DeepCover achieves $$76.7%$$ accuracy, which is $$6%$$ better than the second best extremal." @default.
- W3107312662 created "2020-12-07" @default.
- W3107312662 creator A5020085889 @default.
- W3107312662 creator A5052794441 @default.
- W3107312662 creator A5053852359 @default.
- W3107312662 creator A5077994094 @default.
- W3107312662 date "2020-01-01" @default.
- W3107312662 modified "2023-10-17" @default.
- W3107312662 title "Explaining Image Classifiers Using Statistical Fault Localization" @default.
- W3107312662 cites W2010833880 @default.
- W3107312662 cites W2094837707 @default.
- W3107312662 cites W2101819268 @default.
- W3107312662 cites W2119136132 @default.
- W3107312662 cites W2137819283 @default.
- W3107312662 cites W2144312323 @default.
- W3107312662 cites W2282821441 @default.
- W3107312662 cites W2308882098 @default.
- W3107312662 cites W2510508396 @default.
- W3107312662 cites W2593050976 @default.
- W3107312662 cites W2753783305 @default.
- W3107312662 cites W2792641098 @default.
- W3107312662 cites W2962858109 @default.
- W3107312662 cites W2963913218 @default.
- W3107312662 cites W2988157455 @default.
- W3107312662 cites W3036286896 @default.
- W3107312662 cites W3089623150 @default.
- W3107312662 cites W3091157996 @default.
- W3107312662 cites W4211017739 @default.
- W3107312662 cites W4233482358 @default.
- W3107312662 cites W2796430303 @default.
- W3107312662 doi "https://doi.org/10.1007/978-3-030-58604-1_24" @default.
- W3107312662 hasPublicationYear "2020" @default.
- W3107312662 type Work @default.
- W3107312662 sameAs 3107312662 @default.
- W3107312662 citedByCount "20" @default.
- W3107312662 countsByYear W31073126622020 @default.
- W3107312662 countsByYear W31073126622021 @default.
- W3107312662 countsByYear W31073126622022 @default.
- W3107312662 countsByYear W31073126622023 @default.
- W3107312662 crossrefType "book-chapter" @default.
- W3107312662 hasAuthorship W3107312662A5020085889 @default.
- W3107312662 hasAuthorship W3107312662A5052794441 @default.
- W3107312662 hasAuthorship W3107312662A5053852359 @default.
- W3107312662 hasAuthorship W3107312662A5077994094 @default.
- W3107312662 hasBestOaLocation W31073126622 @default.
- W3107312662 hasConcept C11413529 @default.
- W3107312662 hasConcept C115961682 @default.
- W3107312662 hasConcept C119857082 @default.
- W3107312662 hasConcept C124101348 @default.
- W3107312662 hasConcept C127313418 @default.
- W3107312662 hasConcept C13280743 @default.
- W3107312662 hasConcept C146849305 @default.
- W3107312662 hasConcept C153180895 @default.
- W3107312662 hasConcept C154945302 @default.
- W3107312662 hasConcept C165205528 @default.
- W3107312662 hasConcept C175551986 @default.
- W3107312662 hasConcept C177264268 @default.
- W3107312662 hasConcept C185798385 @default.
- W3107312662 hasConcept C189430467 @default.
- W3107312662 hasConcept C199360897 @default.
- W3107312662 hasConcept C205649164 @default.
- W3107312662 hasConcept C2984842247 @default.
- W3107312662 hasConcept C41008148 @default.
- W3107312662 hasConcept C50644808 @default.
- W3107312662 hasConcept C94966114 @default.
- W3107312662 hasConceptScore W3107312662C11413529 @default.
- W3107312662 hasConceptScore W3107312662C115961682 @default.
- W3107312662 hasConceptScore W3107312662C119857082 @default.
- W3107312662 hasConceptScore W3107312662C124101348 @default.
- W3107312662 hasConceptScore W3107312662C127313418 @default.
- W3107312662 hasConceptScore W3107312662C13280743 @default.
- W3107312662 hasConceptScore W3107312662C146849305 @default.
- W3107312662 hasConceptScore W3107312662C153180895 @default.
- W3107312662 hasConceptScore W3107312662C154945302 @default.
- W3107312662 hasConceptScore W3107312662C165205528 @default.
- W3107312662 hasConceptScore W3107312662C175551986 @default.
- W3107312662 hasConceptScore W3107312662C177264268 @default.
- W3107312662 hasConceptScore W3107312662C185798385 @default.
- W3107312662 hasConceptScore W3107312662C189430467 @default.
- W3107312662 hasConceptScore W3107312662C199360897 @default.
- W3107312662 hasConceptScore W3107312662C205649164 @default.
- W3107312662 hasConceptScore W3107312662C2984842247 @default.
- W3107312662 hasConceptScore W3107312662C41008148 @default.
- W3107312662 hasConceptScore W3107312662C50644808 @default.
- W3107312662 hasConceptScore W3107312662C94966114 @default.
- W3107312662 hasLocation W31073126621 @default.
- W3107312662 hasLocation W31073126622 @default.
- W3107312662 hasLocation W31073126623 @default.
- W3107312662 hasLocation W31073126624 @default.
- W3107312662 hasOpenAccess W3107312662 @default.
- W3107312662 hasPrimaryLocation W31073126621 @default.
- W3107312662 hasRelatedWork W2057836452 @default.
- W3107312662 hasRelatedWork W2373962361 @default.
- W3107312662 hasRelatedWork W2441762250 @default.
- W3107312662 hasRelatedWork W2897842727 @default.
- W3107312662 hasRelatedWork W2939912336 @default.
- W3107312662 hasRelatedWork W2964710935 @default.
- W3107312662 hasRelatedWork W2971533346 @default.