Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312292611> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W4312292611 abstract "Despite rapid advances of modern artificial intelligence (AI), there is a growing concern regarding its capacity to be explainable, transparent, and accountable. One crucial step towards such AI systems involves reliable and efficient uncertainty quantification methods. Existing approaches to uncertainty quantification in natural language processing (NLP) take a Bayesian Deep Learning approach. However, the latter is known to not be computationally efficient in testing time, thus hindering its applicability in real-life scenarios. This paper proposes a new focus on the efficiency of uncertainty quantification methods, evaluating them on four multi-label text classification tasks. Our novel methods of representing epistemic and aleatoric uncertainties enable efficient uncertainty quantification (around 13 to 45 times faster than existing approaches, depending on architecture) with posterior analysis in the (approximated) latent- and data space. We conduct extensive experiments and studies on diverse neural network architectures (LSTM, CNN and Transformer) to analyse their power. Our results prove the benefits of explicitly modelling uncertainty in neural networks." @default.
- W4312292611 created "2023-01-04" @default.
- W4312292611 creator A5003635209 @default.
- W4312292611 creator A5017619842 @default.
- W4312292611 creator A5022110111 @default.
- W4312292611 creator A5034125179 @default.
- W4312292611 creator A5039518853 @default.
- W4312292611 creator A5060294195 @default.
- W4312292611 creator A5083169271 @default.
- W4312292611 date "2022-07-18" @default.
- W4312292611 modified "2023-10-16" @default.
- W4312292611 title "Efficient Uncertainty Quantification for Multilabel Text Classification" @default.
- W4312292611 doi "https://doi.org/10.1109/ijcnn55064.2022.9892871" @default.
- W4312292611 hasPublicationYear "2022" @default.
- W4312292611 type Work @default.
- W4312292611 citedByCount "0" @default.
- W4312292611 crossrefType "proceedings-article" @default.
- W4312292611 hasAuthorship W4312292611A5003635209 @default.
- W4312292611 hasAuthorship W4312292611A5017619842 @default.
- W4312292611 hasAuthorship W4312292611A5022110111 @default.
- W4312292611 hasAuthorship W4312292611A5034125179 @default.
- W4312292611 hasAuthorship W4312292611A5039518853 @default.
- W4312292611 hasAuthorship W4312292611A5060294195 @default.
- W4312292611 hasAuthorship W4312292611A5083169271 @default.
- W4312292611 hasBestOaLocation W43122926112 @default.
- W4312292611 hasConcept C107673813 @default.
- W4312292611 hasConcept C119599485 @default.
- W4312292611 hasConcept C119857082 @default.
- W4312292611 hasConcept C127413603 @default.
- W4312292611 hasConcept C154945302 @default.
- W4312292611 hasConcept C165801399 @default.
- W4312292611 hasConcept C32230216 @default.
- W4312292611 hasConcept C33724603 @default.
- W4312292611 hasConcept C41008148 @default.
- W4312292611 hasConcept C50644808 @default.
- W4312292611 hasConcept C66322947 @default.
- W4312292611 hasConceptScore W4312292611C107673813 @default.
- W4312292611 hasConceptScore W4312292611C119599485 @default.
- W4312292611 hasConceptScore W4312292611C119857082 @default.
- W4312292611 hasConceptScore W4312292611C127413603 @default.
- W4312292611 hasConceptScore W4312292611C154945302 @default.
- W4312292611 hasConceptScore W4312292611C165801399 @default.
- W4312292611 hasConceptScore W4312292611C32230216 @default.
- W4312292611 hasConceptScore W4312292611C33724603 @default.
- W4312292611 hasConceptScore W4312292611C41008148 @default.
- W4312292611 hasConceptScore W4312292611C50644808 @default.
- W4312292611 hasConceptScore W4312292611C66322947 @default.
- W4312292611 hasLocation W43122926111 @default.
- W4312292611 hasLocation W43122926112 @default.
- W4312292611 hasLocation W43122926113 @default.
- W4312292611 hasOpenAccess W4312292611 @default.
- W4312292611 hasPrimaryLocation W43122926111 @default.
- W4312292611 hasRelatedWork W1581446651 @default.
- W4312292611 hasRelatedWork W1956930971 @default.
- W4312292611 hasRelatedWork W2016517455 @default.
- W4312292611 hasRelatedWork W2037915485 @default.
- W4312292611 hasRelatedWork W2364857452 @default.
- W4312292611 hasRelatedWork W2902946190 @default.
- W4312292611 hasRelatedWork W3150265835 @default.
- W4312292611 hasRelatedWork W3154094704 @default.
- W4312292611 hasRelatedWork W32248825 @default.
- W4312292611 hasRelatedWork W1629725936 @default.
- W4312292611 isParatext "false" @default.
- W4312292611 isRetracted "false" @default.
- W4312292611 workType "article" @default.