Matches in SemOpenAlex for { <https://semopenalex.org/work/W3105472275> ?p ?o ?g. }
- W3105472275 abstract "Deep learning classifiers are assisting humans in making decisions and hence the user's trust in these models is of paramount importance. Trust is often a function of constant behavior. From an AI model perspective it means given the same input the user would expect the same output, especially for correct outputs, or in other words consistently correct outputs. This paper studies a model behavior in the context of periodic retraining of deployed models where the outputs from successive generations of the models might not agree on the correct labels assigned to the same input. We formally define consistency and correct-consistency of a learning model. We prove that consistency and correct-consistency of an ensemble learner is not less than the average consistency and correct-consistency of individual learners and correct-consistency can be improved with a probability by combining learners with accuracy not less than the average accuracy of ensemble component learners. To validate the theory using three datasets and two state-of-the-art deep learning classifiers we also propose an efficient dynamic snapshot ensemble method and demonstrate its value." @default.
- W3105472275 created "2020-11-23" @default.
- W3105472275 creator A5008421418 @default.
- W3105472275 creator A5013166935 @default.
- W3105472275 creator A5020293284 @default.
- W3105472275 creator A5030141359 @default.
- W3105472275 creator A5065844089 @default.
- W3105472275 creator A5068587810 @default.
- W3105472275 creator A5072406517 @default.
- W3105472275 creator A5085554484 @default.
- W3105472275 date "2020-11-13" @default.
- W3105472275 modified "2023-10-18" @default.
- W3105472275 title "Wisdom of the Ensemble: Improving Consistency of Deep Learning Models" @default.
- W3105472275 cites W1775434803 @default.
- W3105472275 cites W1797070008 @default.
- W3105472275 cites W1904365287 @default.
- W3105472275 cites W2009727399 @default.
- W3105472275 cites W2108069432 @default.
- W3105472275 cites W2112076978 @default.
- W3105472275 cites W2141125852 @default.
- W3105472275 cites W2141705618 @default.
- W3105472275 cites W2151503710 @default.
- W3105472275 cites W2172734211 @default.
- W3105472275 cites W2194775991 @default.
- W3105472275 cites W2225235391 @default.
- W3105472275 cites W2294567968 @default.
- W3105472275 cites W2313299119 @default.
- W3105472275 cites W2468328197 @default.
- W3105472275 cites W2518108298 @default.
- W3105472275 cites W2555003309 @default.
- W3105472275 cites W2747402019 @default.
- W3105472275 cites W2782812883 @default.
- W3105472275 cites W2787017828 @default.
- W3105472275 cites W2789960424 @default.
- W3105472275 cites W2799424953 @default.
- W3105472275 cites W2801179766 @default.
- W3105472275 cites W2907050536 @default.
- W3105472275 cites W2912934387 @default.
- W3105472275 cites W2951696358 @default.
- W3105472275 cites W2963238274 @default.
- W3105472275 cites W2963286056 @default.
- W3105472275 cites W2964059111 @default.
- W3105472275 cites W2970461927 @default.
- W3105472275 cites W2970859221 @default.
- W3105472275 cites W2992606886 @default.
- W3105472275 cites W3009447076 @default.
- W3105472275 cites W3118608800 @default.
- W3105472275 cites W3120167475 @default.
- W3105472275 doi "https://doi.org/10.48550/arxiv.2011.06796" @default.
- W3105472275 hasPublicationYear "2020" @default.
- W3105472275 type Work @default.
- W3105472275 sameAs 3105472275 @default.
- W3105472275 citedByCount "0" @default.
- W3105472275 crossrefType "posted-content" @default.
- W3105472275 hasAuthorship W3105472275A5008421418 @default.
- W3105472275 hasAuthorship W3105472275A5013166935 @default.
- W3105472275 hasAuthorship W3105472275A5020293284 @default.
- W3105472275 hasAuthorship W3105472275A5030141359 @default.
- W3105472275 hasAuthorship W3105472275A5065844089 @default.
- W3105472275 hasAuthorship W3105472275A5068587810 @default.
- W3105472275 hasAuthorship W3105472275A5072406517 @default.
- W3105472275 hasAuthorship W3105472275A5085554484 @default.
- W3105472275 hasBestOaLocation W31054722751 @default.
- W3105472275 hasConcept C108583219 @default.
- W3105472275 hasConcept C111919701 @default.
- W3105472275 hasConcept C119857082 @default.
- W3105472275 hasConcept C119898033 @default.
- W3105472275 hasConcept C12713177 @default.
- W3105472275 hasConcept C144133560 @default.
- W3105472275 hasConcept C151730666 @default.
- W3105472275 hasConcept C154945302 @default.
- W3105472275 hasConcept C155202549 @default.
- W3105472275 hasConcept C2776436953 @default.
- W3105472275 hasConcept C2778712577 @default.
- W3105472275 hasConcept C2779343474 @default.
- W3105472275 hasConcept C37279795 @default.
- W3105472275 hasConcept C41008148 @default.
- W3105472275 hasConcept C45942800 @default.
- W3105472275 hasConcept C55282118 @default.
- W3105472275 hasConcept C86803240 @default.
- W3105472275 hasConcept C93361087 @default.
- W3105472275 hasConceptScore W3105472275C108583219 @default.
- W3105472275 hasConceptScore W3105472275C111919701 @default.
- W3105472275 hasConceptScore W3105472275C119857082 @default.
- W3105472275 hasConceptScore W3105472275C119898033 @default.
- W3105472275 hasConceptScore W3105472275C12713177 @default.
- W3105472275 hasConceptScore W3105472275C144133560 @default.
- W3105472275 hasConceptScore W3105472275C151730666 @default.
- W3105472275 hasConceptScore W3105472275C154945302 @default.
- W3105472275 hasConceptScore W3105472275C155202549 @default.
- W3105472275 hasConceptScore W3105472275C2776436953 @default.
- W3105472275 hasConceptScore W3105472275C2778712577 @default.
- W3105472275 hasConceptScore W3105472275C2779343474 @default.
- W3105472275 hasConceptScore W3105472275C37279795 @default.
- W3105472275 hasConceptScore W3105472275C41008148 @default.
- W3105472275 hasConceptScore W3105472275C45942800 @default.
- W3105472275 hasConceptScore W3105472275C55282118 @default.
- W3105472275 hasConceptScore W3105472275C86803240 @default.
- W3105472275 hasConceptScore W3105472275C93361087 @default.
- W3105472275 hasLocation W31054722751 @default.