Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287325185> ?p ?o ?g. }
Showing items 1 to 78 of
78
with 100 items per page.
- W4287325185 abstract "Deep neural networks (DNNs) are notorious for making more mistakes for the classes that have substantially fewer samples than the others during training. Such class imbalance is ubiquitous in clinical applications and very crucial to handle because the classes with fewer samples most often correspond to critical cases (e.g., cancer) where misclassifications can have severe consequences. Not to miss such cases, binary classifiers need to be operated at high True Positive Rates (TPRs) by setting a higher threshold, but this comes at the cost of very high False Positive Rates (FPRs) for problems with class imbalance. Existing methods for learning under class imbalance most often do not take this into account. We argue that prediction accuracy should be improved by emphasizing reducing FPRs at high TPRs for problems where misclassification of the positive, i.e. critical, class samples are associated with higher cost. To this end, we pose the training of a DNN for binary classification as a constrained optimization problem and introduce a novel constraint that can be used with existing loss functions to enforce maximal area under the ROC curve (AUC) through prioritizing FPR reduction at high TPR. We solve the resulting constrained optimization problem using an Augmented Lagrangian method (ALM). Going beyond binary, we also propose two possible extensions of the proposed constraint for multi-class classification problems. We present experimental results for image-based binary and multi-class classification applications using an in-house medical imaging dataset, CIFAR10, and CIFAR100. Our results demonstrate that the proposed method improves the baselines in majority of the cases by attaining higher accuracy on critical classes while reducing the misclassification rate for the non-critical class samples." @default.
- W4287325185 created "2022-07-25" @default.
- W4287325185 creator A5036970822 @default.
- W4287325185 creator A5046783553 @default.
- W4287325185 creator A5056195195 @default.
- W4287325185 creator A5075176537 @default.
- W4287325185 creator A5082803116 @default.
- W4287325185 date "2021-02-21" @default.
- W4287325185 modified "2023-10-16" @default.
- W4287325185 title "Constrained Optimization to Train Neural Networks on Critical and Under-Represented Classes" @default.
- W4287325185 doi "https://doi.org/10.48550/arxiv.2102.12894" @default.
- W4287325185 hasPublicationYear "2021" @default.
- W4287325185 type Work @default.
- W4287325185 citedByCount "0" @default.
- W4287325185 crossrefType "posted-content" @default.
- W4287325185 hasAuthorship W4287325185A5036970822 @default.
- W4287325185 hasAuthorship W4287325185A5046783553 @default.
- W4287325185 hasAuthorship W4287325185A5056195195 @default.
- W4287325185 hasAuthorship W4287325185A5075176537 @default.
- W4287325185 hasAuthorship W4287325185A5082803116 @default.
- W4287325185 hasBestOaLocation W42873251851 @default.
- W4287325185 hasConcept C111335779 @default.
- W4287325185 hasConcept C11413529 @default.
- W4287325185 hasConcept C115961682 @default.
- W4287325185 hasConcept C119857082 @default.
- W4287325185 hasConcept C12267149 @default.
- W4287325185 hasConcept C137836250 @default.
- W4287325185 hasConcept C153180895 @default.
- W4287325185 hasConcept C154945302 @default.
- W4287325185 hasConcept C2524010 @default.
- W4287325185 hasConcept C2776036281 @default.
- W4287325185 hasConcept C2777212361 @default.
- W4287325185 hasConcept C2984842247 @default.
- W4287325185 hasConcept C2989514635 @default.
- W4287325185 hasConcept C33923547 @default.
- W4287325185 hasConcept C41008148 @default.
- W4287325185 hasConcept C48372109 @default.
- W4287325185 hasConcept C50644808 @default.
- W4287325185 hasConcept C66905080 @default.
- W4287325185 hasConcept C75294576 @default.
- W4287325185 hasConcept C94375191 @default.
- W4287325185 hasConceptScore W4287325185C111335779 @default.
- W4287325185 hasConceptScore W4287325185C11413529 @default.
- W4287325185 hasConceptScore W4287325185C115961682 @default.
- W4287325185 hasConceptScore W4287325185C119857082 @default.
- W4287325185 hasConceptScore W4287325185C12267149 @default.
- W4287325185 hasConceptScore W4287325185C137836250 @default.
- W4287325185 hasConceptScore W4287325185C153180895 @default.
- W4287325185 hasConceptScore W4287325185C154945302 @default.
- W4287325185 hasConceptScore W4287325185C2524010 @default.
- W4287325185 hasConceptScore W4287325185C2776036281 @default.
- W4287325185 hasConceptScore W4287325185C2777212361 @default.
- W4287325185 hasConceptScore W4287325185C2984842247 @default.
- W4287325185 hasConceptScore W4287325185C2989514635 @default.
- W4287325185 hasConceptScore W4287325185C33923547 @default.
- W4287325185 hasConceptScore W4287325185C41008148 @default.
- W4287325185 hasConceptScore W4287325185C48372109 @default.
- W4287325185 hasConceptScore W4287325185C50644808 @default.
- W4287325185 hasConceptScore W4287325185C66905080 @default.
- W4287325185 hasConceptScore W4287325185C75294576 @default.
- W4287325185 hasConceptScore W4287325185C94375191 @default.
- W4287325185 hasLocation W42873251851 @default.
- W4287325185 hasLocation W42873251852 @default.
- W4287325185 hasOpenAccess W4287325185 @default.
- W4287325185 hasPrimaryLocation W42873251851 @default.
- W4287325185 hasRelatedWork W2029499417 @default.
- W4287325185 hasRelatedWork W2145128058 @default.
- W4287325185 hasRelatedWork W2170850735 @default.
- W4287325185 hasRelatedWork W23263483 @default.
- W4287325185 hasRelatedWork W2355999325 @default.
- W4287325185 hasRelatedWork W2444525338 @default.
- W4287325185 hasRelatedWork W2910954186 @default.
- W4287325185 hasRelatedWork W4205999209 @default.
- W4287325185 hasRelatedWork W4312614571 @default.
- W4287325185 hasRelatedWork W2096295868 @default.
- W4287325185 isParatext "false" @default.
- W4287325185 isRetracted "false" @default.
- W4287325185 workType "article" @default.