Matches in SemOpenAlex for { <https://semopenalex.org/work/W2783994573> ?p ?o ?g. }
- W2783994573 abstract "Errors are the enemy of classification systems, so minimising the total probability of error is an understandable objective in statistical machine learning classifiers. However, for open-world application in trusted autonomous systems , not all errors are equal in terms of their consequences. So, the ability for users and designers to define an objective function that distributes errors according to preference criteria might elevate trust. Previous approaches in cost-sensitive classification have focussed on dealing with distribution imbalances by cost weighting the probability of classification. A novel alternative is proposed that learns a ‘confusion objective’ and is suitable for integration with modular Deep Network architectures. The approach demonstrates an ability to control the error distribution in training of supervised networks via back-propagation for the penalty of an increase in total errors. Theory is developed for the new confusion objective function and compared with cross-entropy and squared loss objectives. The capacity for error shaping is demonstrated via a range of empirical experiments using a shallow and deep network. The classification of handwritten digits from up to three independent databases demonstrates desired error performance is maintained across unforeseen data distributions. Some significant and unique forms of error control are demonstrated and their limitations investigated." @default.
- W2783994573 created "2018-01-26" @default.
- W2783994573 creator A5050703829 @default.
- W2783994573 date "2018-01-01" @default.
- W2783994573 modified "2023-09-29" @default.
- W2783994573 title "Learning to Shape Errors with a Confusion Objective" @default.
- W2783994573 cites W2005530589 @default.
- W2783994573 cites W2048035780 @default.
- W2783994573 cites W2058732827 @default.
- W2783994573 cites W2076063813 @default.
- W2783994573 cites W2103614420 @default.
- W2783994573 cites W2112796928 @default.
- W2783994573 cites W2116444583 @default.
- W2783994573 cites W2141125852 @default.
- W2783994573 cites W4214717370 @default.
- W2783994573 cites W4246836305 @default.
- W2783994573 doi "https://doi.org/10.1007/978-3-319-64816-3_13" @default.
- W2783994573 hasPublicationYear "2018" @default.
- W2783994573 type Work @default.
- W2783994573 sameAs 2783994573 @default.
- W2783994573 citedByCount "0" @default.
- W2783994573 crossrefType "book-chapter" @default.
- W2783994573 hasAuthorship W2783994573A5050703829 @default.
- W2783994573 hasBestOaLocation W27839945731 @default.
- W2783994573 hasConcept C101468663 @default.
- W2783994573 hasConcept C103088060 @default.
- W2783994573 hasConcept C105795698 @default.
- W2783994573 hasConcept C106301342 @default.
- W2783994573 hasConcept C11171543 @default.
- W2783994573 hasConcept C111919701 @default.
- W2783994573 hasConcept C11413529 @default.
- W2783994573 hasConcept C119857082 @default.
- W2783994573 hasConcept C121332964 @default.
- W2783994573 hasConcept C126838900 @default.
- W2783994573 hasConcept C127413603 @default.
- W2783994573 hasConcept C139945424 @default.
- W2783994573 hasConcept C14036430 @default.
- W2783994573 hasConcept C146978453 @default.
- W2783994573 hasConcept C154945302 @default.
- W2783994573 hasConcept C15744967 @default.
- W2783994573 hasConcept C183115368 @default.
- W2783994573 hasConcept C204323151 @default.
- W2783994573 hasConcept C2781140086 @default.
- W2783994573 hasConcept C33923547 @default.
- W2783994573 hasConcept C41008148 @default.
- W2783994573 hasConcept C62520636 @default.
- W2783994573 hasConcept C71924100 @default.
- W2783994573 hasConcept C78458016 @default.
- W2783994573 hasConcept C86803240 @default.
- W2783994573 hasConceptScore W2783994573C101468663 @default.
- W2783994573 hasConceptScore W2783994573C103088060 @default.
- W2783994573 hasConceptScore W2783994573C105795698 @default.
- W2783994573 hasConceptScore W2783994573C106301342 @default.
- W2783994573 hasConceptScore W2783994573C11171543 @default.
- W2783994573 hasConceptScore W2783994573C111919701 @default.
- W2783994573 hasConceptScore W2783994573C11413529 @default.
- W2783994573 hasConceptScore W2783994573C119857082 @default.
- W2783994573 hasConceptScore W2783994573C121332964 @default.
- W2783994573 hasConceptScore W2783994573C126838900 @default.
- W2783994573 hasConceptScore W2783994573C127413603 @default.
- W2783994573 hasConceptScore W2783994573C139945424 @default.
- W2783994573 hasConceptScore W2783994573C14036430 @default.
- W2783994573 hasConceptScore W2783994573C146978453 @default.
- W2783994573 hasConceptScore W2783994573C154945302 @default.
- W2783994573 hasConceptScore W2783994573C15744967 @default.
- W2783994573 hasConceptScore W2783994573C183115368 @default.
- W2783994573 hasConceptScore W2783994573C204323151 @default.
- W2783994573 hasConceptScore W2783994573C2781140086 @default.
- W2783994573 hasConceptScore W2783994573C33923547 @default.
- W2783994573 hasConceptScore W2783994573C41008148 @default.
- W2783994573 hasConceptScore W2783994573C62520636 @default.
- W2783994573 hasConceptScore W2783994573C71924100 @default.
- W2783994573 hasConceptScore W2783994573C78458016 @default.
- W2783994573 hasConceptScore W2783994573C86803240 @default.
- W2783994573 hasLocation W27839945731 @default.
- W2783994573 hasOpenAccess W2783994573 @default.
- W2783994573 hasPrimaryLocation W27839945731 @default.
- W2783994573 hasRelatedWork W120897966 @default.
- W2783994573 hasRelatedWork W1990795458 @default.
- W2783994573 hasRelatedWork W2015363484 @default.
- W2783994573 hasRelatedWork W2042614373 @default.
- W2783994573 hasRelatedWork W2250706843 @default.
- W2783994573 hasRelatedWork W2268813095 @default.
- W2783994573 hasRelatedWork W2736081685 @default.
- W2783994573 hasRelatedWork W2875652255 @default.
- W2783994573 hasRelatedWork W2899682268 @default.
- W2783994573 hasRelatedWork W2918408501 @default.
- W2783994573 hasRelatedWork W3038776261 @default.
- W2783994573 hasRelatedWork W3082570553 @default.
- W2783994573 hasRelatedWork W3091586673 @default.
- W2783994573 hasRelatedWork W3093773656 @default.
- W2783994573 hasRelatedWork W3118640283 @default.
- W2783994573 hasRelatedWork W3129912876 @default.
- W2783994573 hasRelatedWork W3150280470 @default.
- W2783994573 hasRelatedWork W3188049289 @default.
- W2783994573 hasRelatedWork W1606583174 @default.
- W2783994573 hasRelatedWork W287601806 @default.
- W2783994573 isParatext "false" @default.
- W2783994573 isRetracted "false" @default.
- W2783994573 magId "2783994573" @default.