Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204656253> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W3204656253 abstract "Adversarial representation learning aims to learn data representations for a target task while removing unwanted sensitive information at the same time. Existing methods learn model parameters iteratively through stochastic gradient descent-ascent, which is often unstable and unreliable in practice. To overcome this challenge, we adopt closed-form solvers for the adversary and target task. We model them as kernel ridge regressors and analytically determine an upper-bound on the optimal dimensionality of representation. Our solution, dubbed OptNet-ARL, reduces to a stable one one-shot optimization problem that can be solved reliably and efficiently. OptNet-ARL can be easily generalized to the case of multiple target tasks and sensitive attributes. Numerical experiments, on both small and large scale datasets, show that, from an optimization perspective, OptNet-ARL is stable and exhibits three to five times faster convergence. Performance wise, when the target and sensitive attributes are dependent, OptNet-ARL learns representations that offer a better trade-off front between (a) utility and bias for fair classification and (b) utility and privacy by mitigating leakage of private information than existing solutions." @default.
- W3204656253 created "2021-10-11" @default.
- W3204656253 creator A5028859676 @default.
- W3204656253 creator A5031717929 @default.
- W3204656253 creator A5075318218 @default.
- W3204656253 date "2021-01-01" @default.
- W3204656253 modified "2023-09-24" @default.
- W3204656253 title "Adversarial Representation Learning with Closed-Form Solvers" @default.
- W3204656253 cites W1558919105 @default.
- W3204656253 cites W1834627138 @default.
- W3204656253 cites W2018089423 @default.
- W3204656253 cites W2204257188 @default.
- W3204656253 cites W2593768305 @default.
- W3204656253 cites W2888161220 @default.
- W3204656253 cites W2962824803 @default.
- W3204656253 cites W2963070905 @default.
- W3204656253 cites W2963116854 @default.
- W3204656253 cites W2963350032 @default.
- W3204656253 cites W2963394878 @default.
- W3204656253 cites W3009713534 @default.
- W3204656253 cites W3035715279 @default.
- W3204656253 doi "https://doi.org/10.1007/978-3-030-86520-7_45" @default.
- W3204656253 hasPublicationYear "2021" @default.
- W3204656253 type Work @default.
- W3204656253 sameAs 3204656253 @default.
- W3204656253 citedByCount "0" @default.
- W3204656253 crossrefType "book-chapter" @default.
- W3204656253 hasAuthorship W3204656253A5028859676 @default.
- W3204656253 hasAuthorship W3204656253A5031717929 @default.
- W3204656253 hasAuthorship W3204656253A5075318218 @default.
- W3204656253 hasBestOaLocation W32046562532 @default.
- W3204656253 hasConcept C111030470 @default.
- W3204656253 hasConcept C114614502 @default.
- W3204656253 hasConcept C119857082 @default.
- W3204656253 hasConcept C126255220 @default.
- W3204656253 hasConcept C12713177 @default.
- W3204656253 hasConcept C153258448 @default.
- W3204656253 hasConcept C154945302 @default.
- W3204656253 hasConcept C162324750 @default.
- W3204656253 hasConcept C17744445 @default.
- W3204656253 hasConcept C187736073 @default.
- W3204656253 hasConcept C199539241 @default.
- W3204656253 hasConcept C206688291 @default.
- W3204656253 hasConcept C2776359362 @default.
- W3204656253 hasConcept C2777303404 @default.
- W3204656253 hasConcept C2780451532 @default.
- W3204656253 hasConcept C33923547 @default.
- W3204656253 hasConcept C37736160 @default.
- W3204656253 hasConcept C41008148 @default.
- W3204656253 hasConcept C50522688 @default.
- W3204656253 hasConcept C50644808 @default.
- W3204656253 hasConcept C74193536 @default.
- W3204656253 hasConcept C94625758 @default.
- W3204656253 hasConceptScore W3204656253C111030470 @default.
- W3204656253 hasConceptScore W3204656253C114614502 @default.
- W3204656253 hasConceptScore W3204656253C119857082 @default.
- W3204656253 hasConceptScore W3204656253C126255220 @default.
- W3204656253 hasConceptScore W3204656253C12713177 @default.
- W3204656253 hasConceptScore W3204656253C153258448 @default.
- W3204656253 hasConceptScore W3204656253C154945302 @default.
- W3204656253 hasConceptScore W3204656253C162324750 @default.
- W3204656253 hasConceptScore W3204656253C17744445 @default.
- W3204656253 hasConceptScore W3204656253C187736073 @default.
- W3204656253 hasConceptScore W3204656253C199539241 @default.
- W3204656253 hasConceptScore W3204656253C206688291 @default.
- W3204656253 hasConceptScore W3204656253C2776359362 @default.
- W3204656253 hasConceptScore W3204656253C2777303404 @default.
- W3204656253 hasConceptScore W3204656253C2780451532 @default.
- W3204656253 hasConceptScore W3204656253C33923547 @default.
- W3204656253 hasConceptScore W3204656253C37736160 @default.
- W3204656253 hasConceptScore W3204656253C41008148 @default.
- W3204656253 hasConceptScore W3204656253C50522688 @default.
- W3204656253 hasConceptScore W3204656253C50644808 @default.
- W3204656253 hasConceptScore W3204656253C74193536 @default.
- W3204656253 hasConceptScore W3204656253C94625758 @default.
- W3204656253 hasLocation W32046562531 @default.
- W3204656253 hasLocation W32046562532 @default.
- W3204656253 hasOpenAccess W3204656253 @default.
- W3204656253 hasPrimaryLocation W32046562531 @default.
- W3204656253 hasRelatedWork W11023528 @default.
- W3204656253 hasRelatedWork W11144228 @default.
- W3204656253 hasRelatedWork W1243554 @default.
- W3204656253 hasRelatedWork W14789944 @default.
- W3204656253 hasRelatedWork W2700343 @default.
- W3204656253 hasRelatedWork W4085024 @default.
- W3204656253 hasRelatedWork W6147906 @default.
- W3204656253 hasRelatedWork W6908809 @default.
- W3204656253 hasRelatedWork W7177339 @default.
- W3204656253 hasRelatedWork W7382899 @default.
- W3204656253 isParatext "false" @default.
- W3204656253 isRetracted "false" @default.
- W3204656253 magId "3204656253" @default.
- W3204656253 workType "book-chapter" @default.