Matches in SemOpenAlex for { <https://semopenalex.org/work/W3098794533> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W3098794533 endingPage "11332" @default.
- W3098794533 startingPage "11322" @default.
- W3098794533 abstract "Many inference problems in structured prediction can be modeled as maximizing a score function on a space of labels, where graphs are a natural representation to decompose the total score into a sum of unary (nodes) and pairwise (edges) scores. Given a generative model with an undirected connected graph $G$ and true vector of binary labels, it has been previously shown that when $G$ has good expansion properties, such as complete graphs or $d$-regular expanders, one can exactly recover the true labels (with high probability and in polynomial time) from a single noisy observation of each edge and node. We analyze the previously studied generative model by Globerson et al. (2015) under a notion of statistical parity. That is, given a fair binary node labeling, we ask the question whether it is possible to recover the fair assignment, with high probability and in polynomial time, from single edge and node observations. We find that, in contrast to the known trade-offs between fairness and model performance, the addition of the fairness constraint improves the probability of exact recovery. We effectively explain this phenomenon and empirically show how graphs with poor expansion properties, such as grids, are now capable to achieve exact recovery with high probability. Finally, as a byproduct of our analysis, we provide a tighter minimum-eigenvalue bound than that of Weyl's inequality." @default.
- W3098794533 created "2020-11-23" @default.
- W3098794533 creator A5018564606 @default.
- W3098794533 creator A5038965118 @default.
- W3098794533 date "2020-01-01" @default.
- W3098794533 modified "2023-09-24" @default.
- W3098794533 title "Fairness constraints can help exact inference in structured prediction" @default.
- W3098794533 hasPublicationYear "2020" @default.
- W3098794533 type Work @default.
- W3098794533 sameAs 3098794533 @default.
- W3098794533 citedByCount "2" @default.
- W3098794533 countsByYear W30987945332021 @default.
- W3098794533 crossrefType "proceedings-article" @default.
- W3098794533 hasAuthorship W3098794533A5018564606 @default.
- W3098794533 hasAuthorship W3098794533A5038965118 @default.
- W3098794533 hasConcept C105795698 @default.
- W3098794533 hasConcept C118615104 @default.
- W3098794533 hasConcept C121332964 @default.
- W3098794533 hasConcept C127413603 @default.
- W3098794533 hasConcept C14036430 @default.
- W3098794533 hasConcept C154945302 @default.
- W3098794533 hasConcept C158693339 @default.
- W3098794533 hasConcept C184898388 @default.
- W3098794533 hasConcept C2776214188 @default.
- W3098794533 hasConcept C2779982251 @default.
- W3098794533 hasConcept C33923547 @default.
- W3098794533 hasConcept C41008148 @default.
- W3098794533 hasConcept C48372109 @default.
- W3098794533 hasConcept C62520636 @default.
- W3098794533 hasConcept C62611344 @default.
- W3098794533 hasConcept C66938386 @default.
- W3098794533 hasConcept C73555534 @default.
- W3098794533 hasConcept C78023250 @default.
- W3098794533 hasConcept C78458016 @default.
- W3098794533 hasConcept C86803240 @default.
- W3098794533 hasConcept C94375191 @default.
- W3098794533 hasConceptScore W3098794533C105795698 @default.
- W3098794533 hasConceptScore W3098794533C118615104 @default.
- W3098794533 hasConceptScore W3098794533C121332964 @default.
- W3098794533 hasConceptScore W3098794533C127413603 @default.
- W3098794533 hasConceptScore W3098794533C14036430 @default.
- W3098794533 hasConceptScore W3098794533C154945302 @default.
- W3098794533 hasConceptScore W3098794533C158693339 @default.
- W3098794533 hasConceptScore W3098794533C184898388 @default.
- W3098794533 hasConceptScore W3098794533C2776214188 @default.
- W3098794533 hasConceptScore W3098794533C2779982251 @default.
- W3098794533 hasConceptScore W3098794533C33923547 @default.
- W3098794533 hasConceptScore W3098794533C41008148 @default.
- W3098794533 hasConceptScore W3098794533C48372109 @default.
- W3098794533 hasConceptScore W3098794533C62520636 @default.
- W3098794533 hasConceptScore W3098794533C62611344 @default.
- W3098794533 hasConceptScore W3098794533C66938386 @default.
- W3098794533 hasConceptScore W3098794533C73555534 @default.
- W3098794533 hasConceptScore W3098794533C78023250 @default.
- W3098794533 hasConceptScore W3098794533C78458016 @default.
- W3098794533 hasConceptScore W3098794533C86803240 @default.
- W3098794533 hasConceptScore W3098794533C94375191 @default.
- W3098794533 hasLocation W30987945331 @default.
- W3098794533 hasOpenAccess W3098794533 @default.
- W3098794533 hasPrimaryLocation W30987945331 @default.
- W3098794533 hasRelatedWork W1635409103 @default.
- W3098794533 hasRelatedWork W1706338097 @default.
- W3098794533 hasRelatedWork W1876515821 @default.
- W3098794533 hasRelatedWork W1981903670 @default.
- W3098794533 hasRelatedWork W2083970694 @default.
- W3098794533 hasRelatedWork W2104291768 @default.
- W3098794533 hasRelatedWork W2789871079 @default.
- W3098794533 hasRelatedWork W2949501117 @default.
- W3098794533 hasRelatedWork W2950618930 @default.
- W3098794533 hasRelatedWork W2950906236 @default.
- W3098794533 hasRelatedWork W2952793838 @default.
- W3098794533 hasRelatedWork W2971301838 @default.
- W3098794533 hasRelatedWork W3035909655 @default.
- W3098794533 hasRelatedWork W3040504340 @default.
- W3098794533 hasRelatedWork W3046763715 @default.
- W3098794533 hasRelatedWork W3097350880 @default.
- W3098794533 hasRelatedWork W3175065841 @default.
- W3098794533 hasRelatedWork W3200432867 @default.
- W3098794533 hasRelatedWork W2899094596 @default.
- W3098794533 hasRelatedWork W73426805 @default.
- W3098794533 hasVolume "33" @default.
- W3098794533 isParatext "false" @default.
- W3098794533 isRetracted "false" @default.
- W3098794533 magId "3098794533" @default.
- W3098794533 workType "article" @default.