Matches in SemOpenAlex for { <https://semopenalex.org/work/W3202097924> ?p ?o ?g. }
- W3202097924 endingPage "764" @default.
- W3202097924 startingPage "749" @default.
- W3202097924 abstract "In recent years, significant work has been done to include fairness constraints in the training objective of machine learning algorithms. Differently from classical prediction retreatment algorithms, we focus on learning fair representations of the inputs. The challenge is to learn representations that capture most relevant information to predict the targeted output Y, while not containing any information about a sensitive attribute S. We leverage recent work which has been done to estimate the Hirschfeld-Gebelein-Renyi (HGR) maximal correlation coefficient by learning deep neural network transformations and use it as a min-max game to penalize the intrinsic bias in a multi dimensional latent representation. Compared to other dependence measures, the HGR coefficient captures more information about the non-linear dependencies, making the algorithm more efficient in mitigating bias. After providing a theoretical analysis of the consistency of the estimator and its desirable properties for bias mitigation, we empirically study its impact at various levels of neural architectures. We show that acting at intermediate levels of neural architectures provides best expressiveness/generalization abilities for bias mitigation, and that using an HGR based loss is more efficient than more classical adversarial approaches from the literature." @default.
- W3202097924 created "2021-10-11" @default.
- W3202097924 creator A5008031795 @default.
- W3202097924 creator A5027823503 @default.
- W3202097924 creator A5031635996 @default.
- W3202097924 creator A5065404842 @default.
- W3202097924 date "2021-01-01" @default.
- W3202097924 modified "2023-09-25" @default.
- W3202097924 title "Learning Unbiased Representations via Rényi Minimization" @default.
- W3202097924 cites W1883968710 @default.
- W3202097924 cites W1972598934 @default.
- W3202097924 cites W2018582985 @default.
- W3202097924 cites W2026019770 @default.
- W3202097924 cites W2085988980 @default.
- W3202097924 cites W2116984840 @default.
- W3202097924 cites W2120708938 @default.
- W3202097924 cites W2137983211 @default.
- W3202097924 cites W2158108973 @default.
- W3202097924 cites W2170862660 @default.
- W3202097924 cites W2763549966 @default.
- W3202097924 cites W2890680318 @default.
- W3202097924 cites W2892038960 @default.
- W3202097924 cites W2904642422 @default.
- W3202097924 cites W2963100392 @default.
- W3202097924 cites W2963116854 @default.
- W3202097924 cites W2963350032 @default.
- W3202097924 cites W3009713534 @default.
- W3202097924 cites W3034607623 @default.
- W3202097924 cites W3185463067 @default.
- W3202097924 cites W4288617781 @default.
- W3202097924 doi "https://doi.org/10.1007/978-3-030-86520-7_46" @default.
- W3202097924 hasPublicationYear "2021" @default.
- W3202097924 type Work @default.
- W3202097924 sameAs 3202097924 @default.
- W3202097924 citedByCount "2" @default.
- W3202097924 countsByYear W32020979242021 @default.
- W3202097924 countsByYear W32020979242022 @default.
- W3202097924 crossrefType "book-chapter" @default.
- W3202097924 hasAuthorship W3202097924A5008031795 @default.
- W3202097924 hasAuthorship W3202097924A5027823503 @default.
- W3202097924 hasAuthorship W3202097924A5031635996 @default.
- W3202097924 hasAuthorship W3202097924A5065404842 @default.
- W3202097924 hasBestOaLocation W32020979243 @default.
- W3202097924 hasConcept C105795698 @default.
- W3202097924 hasConcept C11413529 @default.
- W3202097924 hasConcept C119857082 @default.
- W3202097924 hasConcept C120665830 @default.
- W3202097924 hasConcept C121332964 @default.
- W3202097924 hasConcept C134306372 @default.
- W3202097924 hasConcept C152139883 @default.
- W3202097924 hasConcept C153083717 @default.
- W3202097924 hasConcept C154945302 @default.
- W3202097924 hasConcept C177148314 @default.
- W3202097924 hasConcept C17744445 @default.
- W3202097924 hasConcept C185429906 @default.
- W3202097924 hasConcept C192209626 @default.
- W3202097924 hasConcept C199539241 @default.
- W3202097924 hasConcept C2776359362 @default.
- W3202097924 hasConcept C2776436953 @default.
- W3202097924 hasConcept C2984842247 @default.
- W3202097924 hasConcept C33923547 @default.
- W3202097924 hasConcept C41008148 @default.
- W3202097924 hasConcept C50644808 @default.
- W3202097924 hasConcept C94625758 @default.
- W3202097924 hasConceptScore W3202097924C105795698 @default.
- W3202097924 hasConceptScore W3202097924C11413529 @default.
- W3202097924 hasConceptScore W3202097924C119857082 @default.
- W3202097924 hasConceptScore W3202097924C120665830 @default.
- W3202097924 hasConceptScore W3202097924C121332964 @default.
- W3202097924 hasConceptScore W3202097924C134306372 @default.
- W3202097924 hasConceptScore W3202097924C152139883 @default.
- W3202097924 hasConceptScore W3202097924C153083717 @default.
- W3202097924 hasConceptScore W3202097924C154945302 @default.
- W3202097924 hasConceptScore W3202097924C177148314 @default.
- W3202097924 hasConceptScore W3202097924C17744445 @default.
- W3202097924 hasConceptScore W3202097924C185429906 @default.
- W3202097924 hasConceptScore W3202097924C192209626 @default.
- W3202097924 hasConceptScore W3202097924C199539241 @default.
- W3202097924 hasConceptScore W3202097924C2776359362 @default.
- W3202097924 hasConceptScore W3202097924C2776436953 @default.
- W3202097924 hasConceptScore W3202097924C2984842247 @default.
- W3202097924 hasConceptScore W3202097924C33923547 @default.
- W3202097924 hasConceptScore W3202097924C41008148 @default.
- W3202097924 hasConceptScore W3202097924C50644808 @default.
- W3202097924 hasConceptScore W3202097924C94625758 @default.
- W3202097924 hasLocation W32020979241 @default.
- W3202097924 hasLocation W32020979242 @default.
- W3202097924 hasLocation W32020979243 @default.
- W3202097924 hasOpenAccess W3202097924 @default.
- W3202097924 hasPrimaryLocation W32020979241 @default.
- W3202097924 hasRelatedWork W1987620537 @default.
- W3202097924 hasRelatedWork W2096696702 @default.
- W3202097924 hasRelatedWork W2797713747 @default.
- W3202097924 hasRelatedWork W2948262272 @default.
- W3202097924 hasRelatedWork W2961085424 @default.
- W3202097924 hasRelatedWork W2963433148 @default.
- W3202097924 hasRelatedWork W3175659316 @default.
- W3202097924 hasRelatedWork W3176808018 @default.