Matches in SemOpenAlex for { <https://semopenalex.org/work/W4280624659> ?p ?o ?g. }
Showing items 1 to 60 of
60
with 100 items per page.
- W4280624659 abstract "Accuracy and individual fairness are both crucial for trustworthy machine learning, but these two aspects are often incompatible with each other so that enhancing one aspect may sacrifice the other inevitably with side effects of true bias or false fairness. We propose in this paper a new fairness criterion, accurate fairness, to align individual fairness with accuracy. Informally, it requires the treatments of an individual and the individual's similar counterparts to conform to a uniform target, i.e., the ground truth of the individual. We prove that accurate fairness also implies typical group fairness criteria over a union of similar sub-populations. We then present a Siamese fairness in-processing approach to minimize the accuracy and fairness losses of a machine learning model under the accurate fairness constraints. To the best of our knowledge, this is the first time that a Siamese approach is adapted for bias mitigation. We also propose fairness confusion matrix-based metrics, fair-precision, fair-recall, and fair-F1 score, to quantify a trade-off between accuracy and individual fairness. Comparative case studies with popular fairness datasets show that our Siamese fairness approach can achieve on average 1.02%-8.78% higher individual fairness (in terms of fairness through awareness) and 8.38%-13.69% higher accuracy, as well as 10.09%-20.57% higher true fair rate, and 5.43%-10.01% higher fair-F1 score, than the state-of-the-art bias mitigation techniques. This demonstrates that our Siamese fairness approach can indeed improve individual fairness without trading accuracy. Finally, the accurate fairness criterion and Siamese fairness approach are applied to mitigate the possible service discrimination with a real Ctrip dataset, by on average fairly serving 112.33% more customers (specifically, 81.29% more customers in an accurately fair way) than baseline models." @default.
- W4280624659 created "2022-05-22" @default.
- W4280624659 creator A5013814252 @default.
- W4280624659 creator A5014451088 @default.
- W4280624659 creator A5065979541 @default.
- W4280624659 date "2022-05-17" @default.
- W4280624659 modified "2023-09-26" @default.
- W4280624659 title "Accurate Fairness: Improving Individual Fairness without Trading Accuracy" @default.
- W4280624659 doi "https://doi.org/10.48550/arxiv.2205.08704" @default.
- W4280624659 hasPublicationYear "2022" @default.
- W4280624659 type Work @default.
- W4280624659 citedByCount "0" @default.
- W4280624659 crossrefType "posted-content" @default.
- W4280624659 hasAuthorship W4280624659A5013814252 @default.
- W4280624659 hasAuthorship W4280624659A5014451088 @default.
- W4280624659 hasAuthorship W4280624659A5065979541 @default.
- W4280624659 hasBestOaLocation W42806246591 @default.
- W4280624659 hasConcept C11413529 @default.
- W4280624659 hasConcept C11867375 @default.
- W4280624659 hasConcept C119857082 @default.
- W4280624659 hasConcept C153701036 @default.
- W4280624659 hasConcept C154945302 @default.
- W4280624659 hasConcept C157764524 @default.
- W4280624659 hasConcept C177972170 @default.
- W4280624659 hasConcept C29202148 @default.
- W4280624659 hasConcept C31258907 @default.
- W4280624659 hasConcept C38652104 @default.
- W4280624659 hasConcept C41008148 @default.
- W4280624659 hasConcept C555944384 @default.
- W4280624659 hasConcept C76155785 @default.
- W4280624659 hasConceptScore W4280624659C11413529 @default.
- W4280624659 hasConceptScore W4280624659C11867375 @default.
- W4280624659 hasConceptScore W4280624659C119857082 @default.
- W4280624659 hasConceptScore W4280624659C153701036 @default.
- W4280624659 hasConceptScore W4280624659C154945302 @default.
- W4280624659 hasConceptScore W4280624659C157764524 @default.
- W4280624659 hasConceptScore W4280624659C177972170 @default.
- W4280624659 hasConceptScore W4280624659C29202148 @default.
- W4280624659 hasConceptScore W4280624659C31258907 @default.
- W4280624659 hasConceptScore W4280624659C38652104 @default.
- W4280624659 hasConceptScore W4280624659C41008148 @default.
- W4280624659 hasConceptScore W4280624659C555944384 @default.
- W4280624659 hasConceptScore W4280624659C76155785 @default.
- W4280624659 hasLocation W42806246591 @default.
- W4280624659 hasLocation W42806246592 @default.
- W4280624659 hasOpenAccess W4280624659 @default.
- W4280624659 hasPrimaryLocation W42806246591 @default.
- W4280624659 hasRelatedWork W2002544160 @default.
- W4280624659 hasRelatedWork W2105353475 @default.
- W4280624659 hasRelatedWork W2381835541 @default.
- W4280624659 hasRelatedWork W2390110963 @default.
- W4280624659 hasRelatedWork W2765588468 @default.
- W4280624659 hasRelatedWork W2961085424 @default.
- W4280624659 hasRelatedWork W4286629047 @default.
- W4280624659 hasRelatedWork W4306321456 @default.
- W4280624659 hasRelatedWork W4306674287 @default.
- W4280624659 hasRelatedWork W4224009465 @default.
- W4280624659 isParatext "false" @default.
- W4280624659 isRetracted "false" @default.
- W4280624659 workType "article" @default.