FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3166444595> ?p ?o ?g. }

• W3166444595 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 (TPR) by setting a higher threshold but this comes at the cost of very high False Positive Rates (FPR) 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.
• W3166444595 created "2021-06-22" @default.
• W3166444595 creator A5036970822 @default.
• W3166444595 creator A5046783553 @default.
• W3166444595 creator A5056195195 @default.
• W3166444595 creator A5075176537 @default.
• W3166444595 creator A5082803116 @default.
• W3166444595 date "2021-02-21" @default.
• W3166444595 modified "2023-09-23" @default.
• W3166444595 title "Constrained Optimization to Train Neural Networks on Critical and Under-Represented Classes" @default.
• W3166444595 cites W1500890010 @default.
• W3166444595 cites W1972978214 @default.
• W3166444595 cites W2086826587 @default.
• W3166444595 cites W2104933073 @default.
• W3166444595 cites W2136903812 @default.
• W3166444595 cites W2142261479 @default.
• W3166444595 cites W2171234124 @default.
• W3166444595 cites W2194775991 @default.
• W3166444595 cites W2328176404 @default.
• W3166444595 cites W2440599146 @default.
• W3166444595 cites W2551637362 @default.
• W3166444595 cites W2592929672 @default.
• W3166444595 cites W2767106145 @default.
• W3166444595 cites W2798766386 @default.
• W3166444595 cites W2914805075 @default.
• W3166444595 cites W2963238274 @default.
• W3166444595 cites W2963351448 @default.
• W3166444595 cites W2963656735 @default.
• W3166444595 cites W2963691377 @default.
• W3166444595 cites W2979849298 @default.
• W3166444595 cites W3014943439 @default.
• W3166444595 cites W3035552357 @default.
• W3166444595 cites W3093667565 @default.
• W3166444595 cites W3105574171 @default.
• W3166444595 cites W3118608800 @default.
• W3166444595 cites W368078881 @default.
• W3166444595 cites W63665094 @default.
• W3166444595 hasPublicationYear "2021" @default.
• W3166444595 type Work @default.
• W3166444595 sameAs 3166444595 @default.
• W3166444595 citedByCount "1" @default.
• W3166444595 countsByYear W31664445952021 @default.
• W3166444595 crossrefType "posted-content" @default.
• W3166444595 hasAuthorship W3166444595A5036970822 @default.
• W3166444595 hasAuthorship W3166444595A5046783553 @default.
• W3166444595 hasAuthorship W3166444595A5056195195 @default.
• W3166444595 hasAuthorship W3166444595A5075176537 @default.
• W3166444595 hasAuthorship W3166444595A5082803116 @default.
• W3166444595 hasConcept C111335779 @default.
• W3166444595 hasConcept C11413529 @default.
• W3166444595 hasConcept C115961682 @default.
• W3166444595 hasConcept C119857082 @default.
• W3166444595 hasConcept C12267149 @default.
• W3166444595 hasConcept C137836250 @default.
• W3166444595 hasConcept C153180895 @default.
• W3166444595 hasConcept C154945302 @default.
• W3166444595 hasConcept C2524010 @default.
• W3166444595 hasConcept C2776036281 @default.
• W3166444595 hasConcept C2777212361 @default.
• W3166444595 hasConcept C2984842247 @default.
• W3166444595 hasConcept C2989514635 @default.
• W3166444595 hasConcept C33923547 @default.
• W3166444595 hasConcept C41008148 @default.
• W3166444595 hasConcept C48372109 @default.
• W3166444595 hasConcept C50644808 @default.
• W3166444595 hasConcept C66905080 @default.
• W3166444595 hasConcept C75294576 @default.
• W3166444595 hasConcept C94375191 @default.
• W3166444595 hasConceptScore W3166444595C111335779 @default.
• W3166444595 hasConceptScore W3166444595C11413529 @default.
• W3166444595 hasConceptScore W3166444595C115961682 @default.
• W3166444595 hasConceptScore W3166444595C119857082 @default.
• W3166444595 hasConceptScore W3166444595C12267149 @default.
• W3166444595 hasConceptScore W3166444595C137836250 @default.
• W3166444595 hasConceptScore W3166444595C153180895 @default.
• W3166444595 hasConceptScore W3166444595C154945302 @default.
• W3166444595 hasConceptScore W3166444595C2524010 @default.
• W3166444595 hasConceptScore W3166444595C2776036281 @default.
• W3166444595 hasConceptScore W3166444595C2777212361 @default.
• W3166444595 hasConceptScore W3166444595C2984842247 @default.
• W3166444595 hasConceptScore W3166444595C2989514635 @default.
• W3166444595 hasConceptScore W3166444595C33923547 @default.
• W3166444595 hasConceptScore W3166444595C41008148 @default.
• W3166444595 hasConceptScore W3166444595C48372109 @default.
• W3166444595 hasConceptScore W3166444595C50644808 @default.
• W3166444595 hasConceptScore W3166444595C66905080 @default.
• W3166444595 hasConceptScore W3166444595C75294576 @default.
• W3166444595 hasConceptScore W3166444595C94375191 @default.
• W3166444595 hasOpenAccess W3166444595 @default.
• W3166444595 hasRelatedWork W1990795458 @default.
• W3166444595 hasRelatedWork W2103768134 @default.
• W3166444595 hasRelatedWork W2114968414 @default.
• W3166444595 hasRelatedWork W2395492705 @default.
• W3166444595 hasRelatedWork W2399563792 @default.
• W3166444595 hasRelatedWork W2759501326 @default.
• W3166444595 hasRelatedWork W2912626019 @default.
• W3166444595 hasRelatedWork W2920156589 @default.
• W3166444595 hasRelatedWork W2924706473 @default.
• W3166444595 hasRelatedWork W3003830418 @default.
• W3166444595 hasRelatedWork W3032632478 @default.

• next