Matches in SemOpenAlex for { <https://semopenalex.org/work/W2899248093> ?p ?o ?g. }
- W2899248093 abstract "We present a generalization of the Cauchy/Lorentzian, Geman-McClure, Welsch/Leclerc, generalized Charbonnier, Charbonnier/pseudo-Huber/L1-L2, and L2 loss functions. By introducing robustness as a continuous parameter, our loss function allows algorithms built around robust loss minimization to be generalized, which improves performance on basic vision tasks such as registration and clustering. Interpreting our loss as the negative log of a univariate density yields a general probability distribution that includes normal and Cauchy distributions as special cases. This probabilistic interpretation enables the training of neural networks in which the robustness of the loss automatically adapts itself during training, which improves performance on learning-based tasks such as generative image synthesis and unsupervised monocular depth estimation, without requiring any manual parameter tuning." @default.
- W2899248093 created "2018-11-09" @default.
- W2899248093 creator A5024117395 @default.
- W2899248093 date "2017-01-11" @default.
- W2899248093 modified "2023-09-27" @default.
- W2899248093 title "A General and Adaptive Robust Loss Function" @default.
- W2899248093 cites W145691285 @default.
- W2899248093 cites W1523985187 @default.
- W2899248093 cites W1776042733 @default.
- W2899248093 cites W1834627138 @default.
- W2899248093 cites W1909320841 @default.
- W2899248093 cites W1959608418 @default.
- W2899248093 cites W2010535094 @default.
- W2899248093 cites W2028207987 @default.
- W2899248093 cites W2031614119 @default.
- W2899248093 cites W2033959528 @default.
- W2899248093 cites W2046033161 @default.
- W2899248093 cites W2058009001 @default.
- W2899248093 cites W2086465792 @default.
- W2899248093 cites W2088616581 @default.
- W2899248093 cites W2099471712 @default.
- W2899248093 cites W2100315781 @default.
- W2899248093 cites W2113945798 @default.
- W2899248093 cites W2121947440 @default.
- W2899248093 cites W2126926806 @default.
- W2899248093 cites W2129793592 @default.
- W2899248093 cites W2132984323 @default.
- W2899248093 cites W2141637993 @default.
- W2899248093 cites W2150066425 @default.
- W2899248093 cites W2164136210 @default.
- W2899248093 cites W2166038949 @default.
- W2899248093 cites W2167034998 @default.
- W2899248093 cites W2750867444 @default.
- W2899248093 cites W2913192828 @default.
- W2899248093 cites W2949634581 @default.
- W2899248093 cites W2963174698 @default.
- W2899248093 cites W2964121744 @default.
- W2899248093 cites W2964167449 @default.
- W2899248093 cites W3005363104 @default.
- W2899248093 cites W44263287 @default.
- W2899248093 hasPublicationYear "2017" @default.
- W2899248093 type Work @default.
- W2899248093 sameAs 2899248093 @default.
- W2899248093 citedByCount "2" @default.
- W2899248093 countsByYear W28992480932019 @default.
- W2899248093 countsByYear W28992480932020 @default.
- W2899248093 crossrefType "posted-content" @default.
- W2899248093 hasAuthorship W2899248093A5024117395 @default.
- W2899248093 hasConcept C104317684 @default.
- W2899248093 hasConcept C105795698 @default.
- W2899248093 hasConcept C11413529 @default.
- W2899248093 hasConcept C126255220 @default.
- W2899248093 hasConcept C134306372 @default.
- W2899248093 hasConcept C147764199 @default.
- W2899248093 hasConcept C154945302 @default.
- W2899248093 hasConcept C177148314 @default.
- W2899248093 hasConcept C185592680 @default.
- W2899248093 hasConcept C28826006 @default.
- W2899248093 hasConcept C33923547 @default.
- W2899248093 hasConcept C41008148 @default.
- W2899248093 hasConcept C49344536 @default.
- W2899248093 hasConcept C49937458 @default.
- W2899248093 hasConcept C50644808 @default.
- W2899248093 hasConcept C55493867 @default.
- W2899248093 hasConcept C63479239 @default.
- W2899248093 hasConcept C65909025 @default.
- W2899248093 hasConcept C73555534 @default.
- W2899248093 hasConceptScore W2899248093C104317684 @default.
- W2899248093 hasConceptScore W2899248093C105795698 @default.
- W2899248093 hasConceptScore W2899248093C11413529 @default.
- W2899248093 hasConceptScore W2899248093C126255220 @default.
- W2899248093 hasConceptScore W2899248093C134306372 @default.
- W2899248093 hasConceptScore W2899248093C147764199 @default.
- W2899248093 hasConceptScore W2899248093C154945302 @default.
- W2899248093 hasConceptScore W2899248093C177148314 @default.
- W2899248093 hasConceptScore W2899248093C185592680 @default.
- W2899248093 hasConceptScore W2899248093C28826006 @default.
- W2899248093 hasConceptScore W2899248093C33923547 @default.
- W2899248093 hasConceptScore W2899248093C41008148 @default.
- W2899248093 hasConceptScore W2899248093C49344536 @default.
- W2899248093 hasConceptScore W2899248093C49937458 @default.
- W2899248093 hasConceptScore W2899248093C50644808 @default.
- W2899248093 hasConceptScore W2899248093C55493867 @default.
- W2899248093 hasConceptScore W2899248093C63479239 @default.
- W2899248093 hasConceptScore W2899248093C65909025 @default.
- W2899248093 hasConceptScore W2899248093C73555534 @default.
- W2899248093 hasLocation W28992480931 @default.
- W2899248093 hasOpenAccess W2899248093 @default.
- W2899248093 hasPrimaryLocation W28992480931 @default.
- W2899248093 hasRelatedWork W1552452485 @default.
- W2899248093 hasRelatedWork W1919402970 @default.
- W2899248093 hasRelatedWork W1974403204 @default.
- W2899248093 hasRelatedWork W2010581677 @default.
- W2899248093 hasRelatedWork W2121508229 @default.
- W2899248093 hasRelatedWork W2510641838 @default.
- W2899248093 hasRelatedWork W2890591861 @default.
- W2899248093 hasRelatedWork W2914055192 @default.
- W2899248093 hasRelatedWork W2986048261 @default.
- W2899248093 hasRelatedWork W3003022452 @default.
- W2899248093 hasRelatedWork W3013293096 @default.