Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287634802> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W4287634802 abstract "Through the use of carefully tailored convolutional neural network architectures, a deep image prior (DIP) can be used to obtain pre-images from latent representation encodings. Though DIP inversion has been known to be superior to conventional regularized inversion strategies such as total variation, such an over-parameterized generator is able to effectively reconstruct even images that are not in the original data distribution. This limitation makes it challenging to utilize such priors for tasks such as counterfactual reasoning, wherein the goal is to generate small, interpretable changes to an image that systematically leads to changes in the model prediction. To this end, we propose a novel regularization strategy based on an auxiliary loss estimator jointly trained with the predictor, which efficiently guides the prior to recover natural pre-images. Our empirical studies with a real-world ISIC skin lesion detection problem clearly evidence the effectiveness of the proposed approach in synthesizing meaningful counterfactuals. In comparison, we find that the standard DIP inversion often proposes visually imperceptible perturbations to irrelevant parts of the image, thus providing no additional insights into the model behavior." @default.
- W4287634802 created "2022-07-25" @default.
- W4287634802 creator A5046632395 @default.
- W4287634802 creator A5074371899 @default.
- W4287634802 creator A5090107273 @default.
- W4287634802 date "2020-10-22" @default.
- W4287634802 modified "2023-09-24" @default.
- W4287634802 title "Using Deep Image Priors to Generate Counterfactual Explanations" @default.
- W4287634802 doi "https://doi.org/10.48550/arxiv.2010.12046" @default.
- W4287634802 hasPublicationYear "2020" @default.
- W4287634802 type Work @default.
- W4287634802 citedByCount "0" @default.
- W4287634802 crossrefType "posted-content" @default.
- W4287634802 hasAuthorship W4287634802A5046632395 @default.
- W4287634802 hasAuthorship W4287634802A5074371899 @default.
- W4287634802 hasAuthorship W4287634802A5090107273 @default.
- W4287634802 hasBestOaLocation W42876348021 @default.
- W4287634802 hasConcept C105795698 @default.
- W4287634802 hasConcept C107673813 @default.
- W4287634802 hasConcept C108650721 @default.
- W4287634802 hasConcept C109007969 @default.
- W4287634802 hasConcept C111472728 @default.
- W4287634802 hasConcept C11413529 @default.
- W4287634802 hasConcept C115961682 @default.
- W4287634802 hasConcept C119857082 @default.
- W4287634802 hasConcept C121332964 @default.
- W4287634802 hasConcept C138885662 @default.
- W4287634802 hasConcept C151730666 @default.
- W4287634802 hasConcept C153180895 @default.
- W4287634802 hasConcept C154945302 @default.
- W4287634802 hasConcept C163258240 @default.
- W4287634802 hasConcept C165464430 @default.
- W4287634802 hasConcept C177769412 @default.
- W4287634802 hasConcept C185429906 @default.
- W4287634802 hasConcept C1893757 @default.
- W4287634802 hasConcept C2776135515 @default.
- W4287634802 hasConcept C2780992000 @default.
- W4287634802 hasConcept C33923547 @default.
- W4287634802 hasConcept C41008148 @default.
- W4287634802 hasConcept C62520636 @default.
- W4287634802 hasConcept C71889745 @default.
- W4287634802 hasConcept C81363708 @default.
- W4287634802 hasConcept C86803240 @default.
- W4287634802 hasConceptScore W4287634802C105795698 @default.
- W4287634802 hasConceptScore W4287634802C107673813 @default.
- W4287634802 hasConceptScore W4287634802C108650721 @default.
- W4287634802 hasConceptScore W4287634802C109007969 @default.
- W4287634802 hasConceptScore W4287634802C111472728 @default.
- W4287634802 hasConceptScore W4287634802C11413529 @default.
- W4287634802 hasConceptScore W4287634802C115961682 @default.
- W4287634802 hasConceptScore W4287634802C119857082 @default.
- W4287634802 hasConceptScore W4287634802C121332964 @default.
- W4287634802 hasConceptScore W4287634802C138885662 @default.
- W4287634802 hasConceptScore W4287634802C151730666 @default.
- W4287634802 hasConceptScore W4287634802C153180895 @default.
- W4287634802 hasConceptScore W4287634802C154945302 @default.
- W4287634802 hasConceptScore W4287634802C163258240 @default.
- W4287634802 hasConceptScore W4287634802C165464430 @default.
- W4287634802 hasConceptScore W4287634802C177769412 @default.
- W4287634802 hasConceptScore W4287634802C185429906 @default.
- W4287634802 hasConceptScore W4287634802C1893757 @default.
- W4287634802 hasConceptScore W4287634802C2776135515 @default.
- W4287634802 hasConceptScore W4287634802C2780992000 @default.
- W4287634802 hasConceptScore W4287634802C33923547 @default.
- W4287634802 hasConceptScore W4287634802C41008148 @default.
- W4287634802 hasConceptScore W4287634802C62520636 @default.
- W4287634802 hasConceptScore W4287634802C71889745 @default.
- W4287634802 hasConceptScore W4287634802C81363708 @default.
- W4287634802 hasConceptScore W4287634802C86803240 @default.
- W4287634802 hasLocation W42876348021 @default.
- W4287634802 hasLocation W42876348022 @default.
- W4287634802 hasOpenAccess W4287634802 @default.
- W4287634802 hasPrimaryLocation W42876348021 @default.
- W4287634802 hasRelatedWork W2175746458 @default.
- W4287634802 hasRelatedWork W2732542196 @default.
- W4287634802 hasRelatedWork W2949457065 @default.
- W4287634802 hasRelatedWork W2955875930 @default.
- W4287634802 hasRelatedWork W2965938919 @default.
- W4287634802 hasRelatedWork W3094465100 @default.
- W4287634802 hasRelatedWork W3202973005 @default.
- W4287634802 hasRelatedWork W4286891671 @default.
- W4287634802 hasRelatedWork W4287634802 @default.
- W4287634802 hasRelatedWork W4287776258 @default.
- W4287634802 isParatext "false" @default.
- W4287634802 isRetracted "false" @default.
- W4287634802 workType "article" @default.