Matches in SemOpenAlex for { <https://semopenalex.org/work/W2978584365> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W2978584365 abstract "Modern deep artificial neural networks have achieved great success in the domain of computer vision and beyond. However, their application to many real-world tasks is undermined by certain limitations, such as overconfident uncertainty estimates on out-of-distribution data or performance deterioration under data distribution shifts. Several types of deep learning models used for density estimation through probabilistic generative modeling have been shown to fail to detect out-of-distribution samples by assigning higher likelihoods to anomalous data. We investigate this failure mode in Variational Autoencoder models, which are also prone to this, and improve upon the out-of-distribution generalization performance of the model by employing an alternative training scheme utilizing negative samples. We present a fully unsupervised version: when the model is trained in an adversarial manner, the generator’s own outputs can be used as negative samples. We demonstrate empirically the effectiveness of the approach in reducing the overconfident likelihood estimates of out-of-distribution inputs on image data." @default.
- W2978584365 created "2019-10-10" @default.
- W2978584365 creator A5015194199 @default.
- W2978584365 creator A5078450617 @default.
- W2978584365 creator A5082032655 @default.
- W2978584365 date "2022-05-16" @default.
- W2978584365 modified "2023-09-29" @default.
- W2978584365 title "Negative Sampling in Variational Autoencoders" @default.
- W2978584365 cites W1959608418 @default.
- W2978584365 cites W2335728318 @default.
- W2978584365 cites W2734358244 @default.
- W2978584365 cites W2750384547 @default.
- W2978584365 cites W2788017640 @default.
- W2978584365 cites W2892375480 @default.
- W2978584365 cites W2902986194 @default.
- W2978584365 cites W2904981516 @default.
- W2978584365 cites W2913300775 @default.
- W2978584365 cites W2948571844 @default.
- W2978584365 cites W2963384319 @default.
- W2978584365 cites W2963546708 @default.
- W2978584365 cites W2963836885 @default.
- W2978584365 cites W2963857374 @default.
- W2978584365 cites W2969609970 @default.
- W2978584365 cites W2980402966 @default.
- W2978584365 cites W3118608800 @default.
- W2978584365 doi "https://doi.org/10.1109/citds54976.2022.9914244" @default.
- W2978584365 hasPublicationYear "2022" @default.
- W2978584365 type Work @default.
- W2978584365 sameAs 2978584365 @default.
- W2978584365 citedByCount "1" @default.
- W2978584365 countsByYear W29785843652020 @default.
- W2978584365 crossrefType "proceedings-article" @default.
- W2978584365 hasAuthorship W2978584365A5015194199 @default.
- W2978584365 hasAuthorship W2978584365A5078450617 @default.
- W2978584365 hasAuthorship W2978584365A5082032655 @default.
- W2978584365 hasBestOaLocation W29785843652 @default.
- W2978584365 hasConcept C101738243 @default.
- W2978584365 hasConcept C106131492 @default.
- W2978584365 hasConcept C108583219 @default.
- W2978584365 hasConcept C119857082 @default.
- W2978584365 hasConcept C121332964 @default.
- W2978584365 hasConcept C134306372 @default.
- W2978584365 hasConcept C140779682 @default.
- W2978584365 hasConcept C153180895 @default.
- W2978584365 hasConcept C154945302 @default.
- W2978584365 hasConcept C163258240 @default.
- W2978584365 hasConcept C177148314 @default.
- W2978584365 hasConcept C19768560 @default.
- W2978584365 hasConcept C2780992000 @default.
- W2978584365 hasConcept C31972630 @default.
- W2978584365 hasConcept C33923547 @default.
- W2978584365 hasConcept C41008148 @default.
- W2978584365 hasConcept C49937458 @default.
- W2978584365 hasConcept C50644808 @default.
- W2978584365 hasConcept C62520636 @default.
- W2978584365 hasConcept C67186912 @default.
- W2978584365 hasConcept C77088390 @default.
- W2978584365 hasConceptScore W2978584365C101738243 @default.
- W2978584365 hasConceptScore W2978584365C106131492 @default.
- W2978584365 hasConceptScore W2978584365C108583219 @default.
- W2978584365 hasConceptScore W2978584365C119857082 @default.
- W2978584365 hasConceptScore W2978584365C121332964 @default.
- W2978584365 hasConceptScore W2978584365C134306372 @default.
- W2978584365 hasConceptScore W2978584365C140779682 @default.
- W2978584365 hasConceptScore W2978584365C153180895 @default.
- W2978584365 hasConceptScore W2978584365C154945302 @default.
- W2978584365 hasConceptScore W2978584365C163258240 @default.
- W2978584365 hasConceptScore W2978584365C177148314 @default.
- W2978584365 hasConceptScore W2978584365C19768560 @default.
- W2978584365 hasConceptScore W2978584365C2780992000 @default.
- W2978584365 hasConceptScore W2978584365C31972630 @default.
- W2978584365 hasConceptScore W2978584365C33923547 @default.
- W2978584365 hasConceptScore W2978584365C41008148 @default.
- W2978584365 hasConceptScore W2978584365C49937458 @default.
- W2978584365 hasConceptScore W2978584365C50644808 @default.
- W2978584365 hasConceptScore W2978584365C62520636 @default.
- W2978584365 hasConceptScore W2978584365C67186912 @default.
- W2978584365 hasConceptScore W2978584365C77088390 @default.
- W2978584365 hasLocation W29785843651 @default.
- W2978584365 hasLocation W29785843652 @default.
- W2978584365 hasLocation W29785843653 @default.
- W2978584365 hasOpenAccess W2978584365 @default.
- W2978584365 hasPrimaryLocation W29785843651 @default.
- W2978584365 hasRelatedWork W2567271240 @default.
- W2978584365 hasRelatedWork W2788487394 @default.
- W2978584365 hasRelatedWork W2922457425 @default.
- W2978584365 hasRelatedWork W2989980351 @default.
- W2978584365 hasRelatedWork W3002526821 @default.
- W2978584365 hasRelatedWork W3044458868 @default.
- W2978584365 hasRelatedWork W4213225422 @default.
- W2978584365 hasRelatedWork W4220775285 @default.
- W2978584365 hasRelatedWork W4250304930 @default.
- W2978584365 hasRelatedWork W4289656111 @default.
- W2978584365 isParatext "false" @default.
- W2978584365 isRetracted "false" @default.
- W2978584365 magId "2978584365" @default.
- W2978584365 workType "article" @default.