Matches in SemOpenAlex for { <https://semopenalex.org/work/W2750585749> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W2750585749 endingPage "6423" @default.
- W2750585749 startingPage "6414" @default.
- W2750585749 abstract "Machine learning is now being used to make crucial decisions about people's lives. For nearly all of these decisions there is a risk that individuals of a certain race, gender, sexual orientation, or any other subpopulation are unfairly discriminated against. Our recent method has demonstrated how to use techniques from counterfactual inference to make predictions fair across different subpopulations. This method requires that one provides the causal model that generated the data at hand. In general, validating all causal implications of the model is not possible without further assumptions. Hence, it is desirable to integrate competing causal models to provide counterfactually fair decisions, regardless of which causal world is the correct one. In this paper, we show how it is possible to make predictions that are approximately fair with respect to multiple possible causal models at once, thus mitigating the problem of exact causal specification. We frame the goal of learning a fair classifier as an optimization problem with fairness constraints entailed by competing causal explanations. We show how this optimization problem can be efficiently solved using gradient-based methods. We demonstrate the flexibility of our model on two real-world fair classification problems. We show that our model can seamlessly balance fairness in multiple worlds with prediction accuracy." @default.
- W2750585749 created "2017-09-15" @default.
- W2750585749 creator A5006180749 @default.
- W2750585749 creator A5008943199 @default.
- W2750585749 creator A5013647396 @default.
- W2750585749 creator A5033941123 @default.
- W2750585749 date "2017-12-04" @default.
- W2750585749 modified "2023-09-27" @default.
- W2750585749 title "When Worlds Collide: Integrating Different Counterfactual Assumptions in Fairness" @default.
- W2750585749 cites W167258127 @default.
- W2750585749 cites W1994407253 @default.
- W2750585749 cites W2100960835 @default.
- W2750585749 cites W2121803403 @default.
- W2750585749 cites W2486285194 @default.
- W2750585749 cites W2582333088 @default.
- W2750585749 cites W2599025709 @default.
- W2750585749 cites W2606171591 @default.
- W2750585749 cites W2753845591 @default.
- W2750585749 cites W2967382200 @default.
- W2750585749 cites W3121588992 @default.
- W2750585749 hasPublicationYear "2017" @default.
- W2750585749 type Work @default.
- W2750585749 sameAs 2750585749 @default.
- W2750585749 citedByCount "44" @default.
- W2750585749 countsByYear W27505857492017 @default.
- W2750585749 countsByYear W27505857492018 @default.
- W2750585749 countsByYear W27505857492019 @default.
- W2750585749 countsByYear W27505857492020 @default.
- W2750585749 countsByYear W27505857492021 @default.
- W2750585749 countsByYear W27505857492022 @default.
- W2750585749 crossrefType "proceedings-article" @default.
- W2750585749 hasAuthorship W2750585749A5006180749 @default.
- W2750585749 hasAuthorship W2750585749A5008943199 @default.
- W2750585749 hasAuthorship W2750585749A5013647396 @default.
- W2750585749 hasAuthorship W2750585749A5033941123 @default.
- W2750585749 hasConcept C105795698 @default.
- W2750585749 hasConcept C108650721 @default.
- W2750585749 hasConcept C11671645 @default.
- W2750585749 hasConcept C119857082 @default.
- W2750585749 hasConcept C149782125 @default.
- W2750585749 hasConcept C154945302 @default.
- W2750585749 hasConcept C15744967 @default.
- W2750585749 hasConcept C158600405 @default.
- W2750585749 hasConcept C2776214188 @default.
- W2750585749 hasConcept C2780598303 @default.
- W2750585749 hasConcept C33923547 @default.
- W2750585749 hasConcept C41008148 @default.
- W2750585749 hasConcept C71889745 @default.
- W2750585749 hasConcept C77805123 @default.
- W2750585749 hasConcept C95623464 @default.
- W2750585749 hasConceptScore W2750585749C105795698 @default.
- W2750585749 hasConceptScore W2750585749C108650721 @default.
- W2750585749 hasConceptScore W2750585749C11671645 @default.
- W2750585749 hasConceptScore W2750585749C119857082 @default.
- W2750585749 hasConceptScore W2750585749C149782125 @default.
- W2750585749 hasConceptScore W2750585749C154945302 @default.
- W2750585749 hasConceptScore W2750585749C15744967 @default.
- W2750585749 hasConceptScore W2750585749C158600405 @default.
- W2750585749 hasConceptScore W2750585749C2776214188 @default.
- W2750585749 hasConceptScore W2750585749C2780598303 @default.
- W2750585749 hasConceptScore W2750585749C33923547 @default.
- W2750585749 hasConceptScore W2750585749C41008148 @default.
- W2750585749 hasConceptScore W2750585749C71889745 @default.
- W2750585749 hasConceptScore W2750585749C77805123 @default.
- W2750585749 hasConceptScore W2750585749C95623464 @default.
- W2750585749 hasLocation W27505857491 @default.
- W2750585749 hasOpenAccess W2750585749 @default.
- W2750585749 hasPrimaryLocation W27505857491 @default.
- W2750585749 hasRelatedWork W1961345416 @default.
- W2750585749 hasRelatedWork W2014352947 @default.
- W2750585749 hasRelatedWork W2097246321 @default.
- W2750585749 hasRelatedWork W2100960835 @default.
- W2750585749 hasRelatedWork W2116984840 @default.
- W2750585749 hasRelatedWork W2143891888 @default.
- W2750585749 hasRelatedWork W2162670686 @default.
- W2750585749 hasRelatedWork W2530395818 @default.
- W2750585749 hasRelatedWork W2540757487 @default.
- W2750585749 hasRelatedWork W2584805976 @default.
- W2750585749 hasRelatedWork W2753845591 @default.
- W2750585749 hasRelatedWork W2809878087 @default.
- W2750585749 hasRelatedWork W2962977061 @default.
- W2750585749 hasRelatedWork W2963174898 @default.
- W2750585749 hasRelatedWork W2963453196 @default.
- W2750585749 hasRelatedWork W2964031043 @default.
- W2750585749 hasRelatedWork W2964060106 @default.
- W2750585749 hasRelatedWork W3023309920 @default.
- W2750585749 hasRelatedWork W3120740533 @default.
- W2750585749 hasRelatedWork W3123374861 @default.
- W2750585749 hasVolume "30" @default.
- W2750585749 isParatext "false" @default.
- W2750585749 isRetracted "false" @default.
- W2750585749 magId "2750585749" @default.
- W2750585749 workType "article" @default.