Matches in SemOpenAlex for { <https://semopenalex.org/work/W4297631911> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W4297631911 abstract "ProBoost, a new boosting algorithm for probabilistic classifiers, is proposed in this work. This algorithm uses the epistemic uncertainty of each training sample to determine the most challenging/uncertain ones; the relevance of these samples is then increased for the next weak learner, producing a sequence that progressively focuses on the samples found to have the highest uncertainty. In the end, the weak learners' outputs are combined into a weighted ensemble of classifiers. Three methods are proposed to manipulate the training set: undersampling, oversampling, and weighting the training samples according to the uncertainty estimated by the weak learners. Furthermore, two approaches are studied regarding the ensemble combination. The weak learner herein considered is a standard convolutional neural network, and the probabilistic models underlying the uncertainty estimation use either variational inference or Monte Carlo dropout. The experimental evaluation carried out on MNIST benchmark datasets shows that ProBoost yields a significant performance improvement. The results are further highlighted by assessing the relative achievable improvement, a metric proposed in this work, which shows that a model with only four weak learners leads to an improvement exceeding 12% in this metric (for either accuracy, sensitivity, or specificity), in comparison to the model learned without ProBoost." @default.
- W4297631911 created "2022-09-30" @default.
- W4297631911 creator A5026826555 @default.
- W4297631911 creator A5039834779 @default.
- W4297631911 creator A5046020340 @default.
- W4297631911 creator A5062543897 @default.
- W4297631911 creator A5083968722 @default.
- W4297631911 date "2022-09-04" @default.
- W4297631911 modified "2023-09-25" @default.
- W4297631911 title "ProBoost: a Boosting Method for Probabilistic Classifiers" @default.
- W4297631911 doi "https://doi.org/10.48550/arxiv.2209.01611" @default.
- W4297631911 hasPublicationYear "2022" @default.
- W4297631911 type Work @default.
- W4297631911 citedByCount "0" @default.
- W4297631911 crossrefType "posted-content" @default.
- W4297631911 hasAuthorship W4297631911A5026826555 @default.
- W4297631911 hasAuthorship W4297631911A5039834779 @default.
- W4297631911 hasAuthorship W4297631911A5046020340 @default.
- W4297631911 hasAuthorship W4297631911A5062543897 @default.
- W4297631911 hasAuthorship W4297631911A5083968722 @default.
- W4297631911 hasBestOaLocation W42976319111 @default.
- W4297631911 hasConcept C105795698 @default.
- W4297631911 hasConcept C119857082 @default.
- W4297631911 hasConcept C126838900 @default.
- W4297631911 hasConcept C127413603 @default.
- W4297631911 hasConcept C136536468 @default.
- W4297631911 hasConcept C150921843 @default.
- W4297631911 hasConcept C154945302 @default.
- W4297631911 hasConcept C176217482 @default.
- W4297631911 hasConcept C183115368 @default.
- W4297631911 hasConcept C185429906 @default.
- W4297631911 hasConcept C190502265 @default.
- W4297631911 hasConcept C197323446 @default.
- W4297631911 hasConcept C21547014 @default.
- W4297631911 hasConcept C2776214188 @default.
- W4297631911 hasConcept C2776257435 @default.
- W4297631911 hasConcept C31258907 @default.
- W4297631911 hasConcept C33923547 @default.
- W4297631911 hasConcept C41008148 @default.
- W4297631911 hasConcept C45942800 @default.
- W4297631911 hasConcept C46686674 @default.
- W4297631911 hasConcept C49937458 @default.
- W4297631911 hasConcept C50644808 @default.
- W4297631911 hasConcept C71924100 @default.
- W4297631911 hasConcept C81790035 @default.
- W4297631911 hasConceptScore W4297631911C105795698 @default.
- W4297631911 hasConceptScore W4297631911C119857082 @default.
- W4297631911 hasConceptScore W4297631911C126838900 @default.
- W4297631911 hasConceptScore W4297631911C127413603 @default.
- W4297631911 hasConceptScore W4297631911C136536468 @default.
- W4297631911 hasConceptScore W4297631911C150921843 @default.
- W4297631911 hasConceptScore W4297631911C154945302 @default.
- W4297631911 hasConceptScore W4297631911C176217482 @default.
- W4297631911 hasConceptScore W4297631911C183115368 @default.
- W4297631911 hasConceptScore W4297631911C185429906 @default.
- W4297631911 hasConceptScore W4297631911C190502265 @default.
- W4297631911 hasConceptScore W4297631911C197323446 @default.
- W4297631911 hasConceptScore W4297631911C21547014 @default.
- W4297631911 hasConceptScore W4297631911C2776214188 @default.
- W4297631911 hasConceptScore W4297631911C2776257435 @default.
- W4297631911 hasConceptScore W4297631911C31258907 @default.
- W4297631911 hasConceptScore W4297631911C33923547 @default.
- W4297631911 hasConceptScore W4297631911C41008148 @default.
- W4297631911 hasConceptScore W4297631911C45942800 @default.
- W4297631911 hasConceptScore W4297631911C46686674 @default.
- W4297631911 hasConceptScore W4297631911C49937458 @default.
- W4297631911 hasConceptScore W4297631911C50644808 @default.
- W4297631911 hasConceptScore W4297631911C71924100 @default.
- W4297631911 hasConceptScore W4297631911C81790035 @default.
- W4297631911 hasLocation W42976319111 @default.
- W4297631911 hasOpenAccess W4297631911 @default.
- W4297631911 hasPrimaryLocation W42976319111 @default.
- W4297631911 hasRelatedWork W1971057815 @default.
- W4297631911 hasRelatedWork W2042086215 @default.
- W4297631911 hasRelatedWork W2891710018 @default.
- W4297631911 hasRelatedWork W3092278000 @default.
- W4297631911 hasRelatedWork W3138490155 @default.
- W4297631911 hasRelatedWork W3193372619 @default.
- W4297631911 hasRelatedWork W3199668957 @default.
- W4297631911 hasRelatedWork W4288758312 @default.
- W4297631911 hasRelatedWork W4297631911 @default.
- W4297631911 hasRelatedWork W4322743207 @default.
- W4297631911 isParatext "false" @default.
- W4297631911 isRetracted "false" @default.
- W4297631911 workType "article" @default.