Matches in SemOpenAlex for { <https://semopenalex.org/work/W3206702512> ?p ?o ?g. }
- W3206702512 endingPage "3359" @default.
- W3206702512 startingPage "3333" @default.
- W3206702512 abstract "Many methods to explain black-box models, whether local or global, are additive. In this paper, we study global additive explanations for non-additive models, focusing on four explanation methods: partial dependence, Shapley explanations adapted to a global setting, distilled additive explanations, and gradient-based explanations. We show that different explanation methods characterize non-additive components in a black-box model’s prediction function in different ways. We use the concepts of main and total effects to anchor additive explanations, and quantitatively evaluate additive and non-additive explanations. Even though distilled explanations are generally the most accurate additive explanations, non-additive explanations such as tree explanations that explicitly model non-additive components tend to be even more accurate. Despite this, our user study showed that machine learning practitioners were better able to leverage additive explanations for various tasks. These considerations should be taken into account when considering which explanation to trust and use to explain black-box models." @default.
- W3206702512 created "2021-10-25" @default.
- W3206702512 creator A5042959700 @default.
- W3206702512 creator A5052163495 @default.
- W3206702512 creator A5053497867 @default.
- W3206702512 creator A5078275541 @default.
- W3206702512 creator A5083893060 @default.
- W3206702512 date "2023-06-19" @default.
- W3206702512 modified "2023-10-02" @default.
- W3206702512 title "Considerations when learning additive explanations for black-box models" @default.
- W3206702512 cites W1589072770 @default.
- W3206702512 cites W1630723770 @default.
- W3206702512 cites W1678356000 @default.
- W3206702512 cites W1787224781 @default.
- W3206702512 cites W1996796871 @default.
- W3206702512 cites W2046945713 @default.
- W3206702512 cites W2048231652 @default.
- W3206702512 cites W2084701050 @default.
- W3206702512 cites W2088203270 @default.
- W3206702512 cites W2094839437 @default.
- W3206702512 cites W2118022153 @default.
- W3206702512 cites W2129888542 @default.
- W3206702512 cites W2149033360 @default.
- W3206702512 cites W2149426898 @default.
- W3206702512 cites W2282821441 @default.
- W3206702512 cites W2294370754 @default.
- W3206702512 cites W2657631929 @default.
- W3206702512 cites W2744365997 @default.
- W3206702512 cites W2787894218 @default.
- W3206702512 cites W2910028250 @default.
- W3206702512 cites W2945976633 @default.
- W3206702512 cites W2948056727 @default.
- W3206702512 cites W2959587146 @default.
- W3206702512 cites W2963399068 @default.
- W3206702512 cites W2963749936 @default.
- W3206702512 cites W2972854536 @default.
- W3206702512 cites W2997560917 @default.
- W3206702512 cites W3005086430 @default.
- W3206702512 cites W3016099278 @default.
- W3206702512 cites W3035264434 @default.
- W3206702512 cites W3035517615 @default.
- W3206702512 cites W3094294665 @default.
- W3206702512 cites W3120388949 @default.
- W3206702512 cites W3121562452 @default.
- W3206702512 cites W3134080822 @default.
- W3206702512 cites W3163411042 @default.
- W3206702512 cites W3171322015 @default.
- W3206702512 cites W3176033544 @default.
- W3206702512 cites W4289097366 @default.
- W3206702512 cites W4298870098 @default.
- W3206702512 doi "https://doi.org/10.1007/s10994-023-06335-8" @default.
- W3206702512 hasPublicationYear "2023" @default.
- W3206702512 type Work @default.
- W3206702512 sameAs 3206702512 @default.
- W3206702512 citedByCount "3" @default.
- W3206702512 countsByYear W32067025122021 @default.
- W3206702512 countsByYear W32067025122023 @default.
- W3206702512 crossrefType "journal-article" @default.
- W3206702512 hasAuthorship W3206702512A5042959700 @default.
- W3206702512 hasAuthorship W3206702512A5052163495 @default.
- W3206702512 hasAuthorship W3206702512A5053497867 @default.
- W3206702512 hasAuthorship W3206702512A5078275541 @default.
- W3206702512 hasAuthorship W3206702512A5083893060 @default.
- W3206702512 hasBestOaLocation W32067025122 @default.
- W3206702512 hasConcept C119857082 @default.
- W3206702512 hasConcept C134306372 @default.
- W3206702512 hasConcept C14036430 @default.
- W3206702512 hasConcept C149782125 @default.
- W3206702512 hasConcept C153083717 @default.
- W3206702512 hasConcept C154945302 @default.
- W3206702512 hasConcept C194648359 @default.
- W3206702512 hasConcept C203223496 @default.
- W3206702512 hasConcept C33923547 @default.
- W3206702512 hasConcept C38803922 @default.
- W3206702512 hasConcept C41008148 @default.
- W3206702512 hasConcept C78458016 @default.
- W3206702512 hasConcept C86803240 @default.
- W3206702512 hasConcept C94966114 @default.
- W3206702512 hasConceptScore W3206702512C119857082 @default.
- W3206702512 hasConceptScore W3206702512C134306372 @default.
- W3206702512 hasConceptScore W3206702512C14036430 @default.
- W3206702512 hasConceptScore W3206702512C149782125 @default.
- W3206702512 hasConceptScore W3206702512C153083717 @default.
- W3206702512 hasConceptScore W3206702512C154945302 @default.
- W3206702512 hasConceptScore W3206702512C194648359 @default.
- W3206702512 hasConceptScore W3206702512C203223496 @default.
- W3206702512 hasConceptScore W3206702512C33923547 @default.
- W3206702512 hasConceptScore W3206702512C38803922 @default.
- W3206702512 hasConceptScore W3206702512C41008148 @default.
- W3206702512 hasConceptScore W3206702512C78458016 @default.
- W3206702512 hasConceptScore W3206702512C86803240 @default.
- W3206702512 hasConceptScore W3206702512C94966114 @default.
- W3206702512 hasFunder F4320335353 @default.
- W3206702512 hasIssue "9" @default.
- W3206702512 hasLocation W32067025121 @default.
- W3206702512 hasLocation W32067025122 @default.
- W3206702512 hasOpenAccess W3206702512 @default.
- W3206702512 hasPrimaryLocation W32067025121 @default.