Matches in SemOpenAlex for { <https://semopenalex.org/work/W2899682268> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W2899682268 abstract "Recent advances in deep learning have achieved impressive gains in classification accuracy on a variety of types of data, including images and text. Despite these gains, however, concerns have been raised about the calibration, robustness, and interpretability of these models. In this paper we propose a simple way to modify any conventional deep architecture to automatically provide more transparent explanations for classification decisions, as well as an intuitive notion of the credibility of each prediction. Specifically, we draw on ideas from nonparametric kernel regression, and propose to predict labels based on a weighted sum of training instances, where the weights are determined by distance in a learned instance-embedding space. Working within the framework of conformal methods, we propose a new measure of nonconformity suggested by our model, and experimentally validate the accompanying theoretical expectations, demonstrating improved transparency, controlled error rates, and robustness to out-of-domain data, without compromising on accuracy or calibration." @default.
- W2899682268 created "2018-11-16" @default.
- W2899682268 creator A5006657942 @default.
- W2899682268 creator A5070767948 @default.
- W2899682268 creator A5088517824 @default.
- W2899682268 date "2019-01-29" @default.
- W2899682268 modified "2023-10-05" @default.
- W2899682268 title "Deep Weighted Averaging Classifiers" @default.
- W2899682268 cites W1932198206 @default.
- W2899682268 cites W1994606570 @default.
- W2899682268 cites W2013587512 @default.
- W2899682268 cites W2027731328 @default.
- W2899682268 cites W2085828897 @default.
- W2899682268 cites W2087347434 @default.
- W2899682268 cites W2114524997 @default.
- W2899682268 cites W2119821739 @default.
- W2899682268 cites W2121949863 @default.
- W2899682268 cites W2122111042 @default.
- W2899682268 cites W2135046866 @default.
- W2899682268 cites W2149033360 @default.
- W2899682268 cites W2157364932 @default.
- W2899682268 cites W2164878629 @default.
- W2899682268 cites W2169495281 @default.
- W2899682268 cites W2251410829 @default.
- W2899682268 cites W2282821441 @default.
- W2899682268 cites W2367397349 @default.
- W2899682268 cites W2799007037 @default.
- W2899682268 cites W2963223515 @default.
- W2899682268 cites W2963266575 @default.
- W2899682268 cites W2964142373 @default.
- W2899682268 cites W2997476292 @default.
- W2899682268 cites W4230674625 @default.
- W2899682268 doi "https://doi.org/10.1145/3287560.3287595" @default.
- W2899682268 hasPublicationYear "2019" @default.
- W2899682268 type Work @default.
- W2899682268 sameAs 2899682268 @default.
- W2899682268 citedByCount "21" @default.
- W2899682268 countsByYear W28996822682019 @default.
- W2899682268 countsByYear W28996822682020 @default.
- W2899682268 countsByYear W28996822682021 @default.
- W2899682268 countsByYear W28996822682022 @default.
- W2899682268 countsByYear W28996822682023 @default.
- W2899682268 crossrefType "proceedings-article" @default.
- W2899682268 hasAuthorship W2899682268A5006657942 @default.
- W2899682268 hasAuthorship W2899682268A5070767948 @default.
- W2899682268 hasAuthorship W2899682268A5088517824 @default.
- W2899682268 hasBestOaLocation W28996822681 @default.
- W2899682268 hasConcept C104317684 @default.
- W2899682268 hasConcept C108583219 @default.
- W2899682268 hasConcept C119857082 @default.
- W2899682268 hasConcept C124101348 @default.
- W2899682268 hasConcept C153180895 @default.
- W2899682268 hasConcept C154945302 @default.
- W2899682268 hasConcept C17744445 @default.
- W2899682268 hasConcept C185592680 @default.
- W2899682268 hasConcept C199539241 @default.
- W2899682268 hasConcept C2780224610 @default.
- W2899682268 hasConcept C2781067378 @default.
- W2899682268 hasConcept C41008148 @default.
- W2899682268 hasConcept C41608201 @default.
- W2899682268 hasConcept C55493867 @default.
- W2899682268 hasConcept C63479239 @default.
- W2899682268 hasConcept C95623464 @default.
- W2899682268 hasConceptScore W2899682268C104317684 @default.
- W2899682268 hasConceptScore W2899682268C108583219 @default.
- W2899682268 hasConceptScore W2899682268C119857082 @default.
- W2899682268 hasConceptScore W2899682268C124101348 @default.
- W2899682268 hasConceptScore W2899682268C153180895 @default.
- W2899682268 hasConceptScore W2899682268C154945302 @default.
- W2899682268 hasConceptScore W2899682268C17744445 @default.
- W2899682268 hasConceptScore W2899682268C185592680 @default.
- W2899682268 hasConceptScore W2899682268C199539241 @default.
- W2899682268 hasConceptScore W2899682268C2780224610 @default.
- W2899682268 hasConceptScore W2899682268C2781067378 @default.
- W2899682268 hasConceptScore W2899682268C41008148 @default.
- W2899682268 hasConceptScore W2899682268C41608201 @default.
- W2899682268 hasConceptScore W2899682268C55493867 @default.
- W2899682268 hasConceptScore W2899682268C63479239 @default.
- W2899682268 hasConceptScore W2899682268C95623464 @default.
- W2899682268 hasFunder F4320306076 @default.
- W2899682268 hasLocation W28996822681 @default.
- W2899682268 hasLocation W28996822682 @default.
- W2899682268 hasLocation W28996822683 @default.
- W2899682268 hasOpenAccess W2899682268 @default.
- W2899682268 hasPrimaryLocation W28996822681 @default.
- W2899682268 hasRelatedWork W3006943036 @default.
- W2899682268 hasRelatedWork W3129898729 @default.
- W2899682268 hasRelatedWork W4200027074 @default.
- W2899682268 hasRelatedWork W4206534706 @default.
- W2899682268 hasRelatedWork W4229079080 @default.
- W2899682268 hasRelatedWork W4299487748 @default.
- W2899682268 hasRelatedWork W4310880831 @default.
- W2899682268 hasRelatedWork W4380075502 @default.
- W2899682268 hasRelatedWork W4385957992 @default.
- W2899682268 hasRelatedWork W4385965371 @default.
- W2899682268 isParatext "false" @default.
- W2899682268 isRetracted "false" @default.
- W2899682268 magId "2899682268" @default.
- W2899682268 workType "article" @default.