Matches in SemOpenAlex for { <https://semopenalex.org/work/W2917263758> ?p ?o ?g. }
- W2917263758 abstract "Most of the work on interpretable machine learning has focused on designing either inherently interpretable models, which typically trade-off accuracy for interpretability, or post-hoc explanation systems, whose explanation quality can be unpredictable. Our method, ExpO, is a hybridization of these approaches that regularizes a model for explanation quality at training time. Importantly, these regularizers are differentiable, model agnostic, and require no domain knowledge to define. We demonstrate that post-hoc explanations for ExpO-regularized models have better explanation quality, as measured by the common fidelity and stability metrics. We verify that improving these metrics leads to significantly more useful explanations with a user study on a realistic task." @default.
- W2917263758 created "2019-03-02" @default.
- W2917263758 creator A5008835202 @default.
- W2917263758 creator A5009547049 @default.
- W2917263758 creator A5029768722 @default.
- W2917263758 creator A5033668205 @default.
- W2917263758 creator A5053097987 @default.
- W2917263758 creator A5060162507 @default.
- W2917263758 date "2019-02-18" @default.
- W2917263758 modified "2023-09-27" @default.
- W2917263758 title "Regularizing Black-box Models for Improved Interpretability" @default.
- W2917263758 cites W1522301498 @default.
- W2917263758 cites W1849277567 @default.
- W2917263758 cites W1996796871 @default.
- W2917263758 cites W2282821441 @default.
- W2917263758 cites W2342045095 @default.
- W2917263758 cites W2346578521 @default.
- W2917263758 cites W2411881509 @default.
- W2917263758 cites W2439568532 @default.
- W2917263758 cites W2440159969 @default.
- W2917263758 cites W2594475271 @default.
- W2917263758 cites W2619325105 @default.
- W2917263758 cites W2626639386 @default.
- W2917263758 cites W2766912318 @default.
- W2917263758 cites W2788403449 @default.
- W2917263758 cites W2809283485 @default.
- W2917263758 cites W2809671526 @default.
- W2917263758 cites W2810120222 @default.
- W2917263758 cites W2891612330 @default.
- W2917263758 cites W2898138693 @default.
- W2917263758 cites W2917231671 @default.
- W2917263758 cites W2949197630 @default.
- W2917263758 cites W2962851944 @default.
- W2917263758 cites W2962858109 @default.
- W2917263758 cites W2962862931 @default.
- W2917263758 cites W2963483561 @default.
- W2917263758 cites W2963798744 @default.
- W2917263758 cites W2964134873 @default.
- W2917263758 cites W2964244352 @default.
- W2917263758 cites W2968060152 @default.
- W2917263758 cites W2970049488 @default.
- W2917263758 cites W2975074147 @default.
- W2917263758 cites W2992167130 @default.
- W2917263758 cites W3004193104 @default.
- W2917263758 cites W3005073185 @default.
- W2917263758 cites W3104674222 @default.
- W2917263758 cites W3105630949 @default.
- W2917263758 hasPublicationYear "2019" @default.
- W2917263758 type Work @default.
- W2917263758 sameAs 2917263758 @default.
- W2917263758 citedByCount "11" @default.
- W2917263758 countsByYear W29172637582019 @default.
- W2917263758 countsByYear W29172637582020 @default.
- W2917263758 countsByYear W29172637582021 @default.
- W2917263758 countsByYear W29172637582022 @default.
- W2917263758 crossrefType "posted-content" @default.
- W2917263758 hasAuthorship W2917263758A5008835202 @default.
- W2917263758 hasAuthorship W2917263758A5009547049 @default.
- W2917263758 hasAuthorship W2917263758A5029768722 @default.
- W2917263758 hasAuthorship W2917263758A5033668205 @default.
- W2917263758 hasAuthorship W2917263758A5053097987 @default.
- W2917263758 hasAuthorship W2917263758A5060162507 @default.
- W2917263758 hasConcept C111472728 @default.
- W2917263758 hasConcept C112972136 @default.
- W2917263758 hasConcept C119857082 @default.
- W2917263758 hasConcept C134306372 @default.
- W2917263758 hasConcept C138885662 @default.
- W2917263758 hasConcept C154945302 @default.
- W2917263758 hasConcept C162324750 @default.
- W2917263758 hasConcept C187736073 @default.
- W2917263758 hasConcept C199343813 @default.
- W2917263758 hasConcept C202615002 @default.
- W2917263758 hasConcept C2776459999 @default.
- W2917263758 hasConcept C2779530757 @default.
- W2917263758 hasConcept C2780451532 @default.
- W2917263758 hasConcept C2781067378 @default.
- W2917263758 hasConcept C2992886853 @default.
- W2917263758 hasConcept C33923547 @default.
- W2917263758 hasConcept C36503486 @default.
- W2917263758 hasConcept C41008148 @default.
- W2917263758 hasConcept C71924100 @default.
- W2917263758 hasConcept C76155785 @default.
- W2917263758 hasConcept C94966114 @default.
- W2917263758 hasConceptScore W2917263758C111472728 @default.
- W2917263758 hasConceptScore W2917263758C112972136 @default.
- W2917263758 hasConceptScore W2917263758C119857082 @default.
- W2917263758 hasConceptScore W2917263758C134306372 @default.
- W2917263758 hasConceptScore W2917263758C138885662 @default.
- W2917263758 hasConceptScore W2917263758C154945302 @default.
- W2917263758 hasConceptScore W2917263758C162324750 @default.
- W2917263758 hasConceptScore W2917263758C187736073 @default.
- W2917263758 hasConceptScore W2917263758C199343813 @default.
- W2917263758 hasConceptScore W2917263758C202615002 @default.
- W2917263758 hasConceptScore W2917263758C2776459999 @default.
- W2917263758 hasConceptScore W2917263758C2779530757 @default.
- W2917263758 hasConceptScore W2917263758C2780451532 @default.
- W2917263758 hasConceptScore W2917263758C2781067378 @default.
- W2917263758 hasConceptScore W2917263758C2992886853 @default.
- W2917263758 hasConceptScore W2917263758C33923547 @default.
- W2917263758 hasConceptScore W2917263758C36503486 @default.