Matches in SemOpenAlex for { <https://semopenalex.org/work/W3159348040> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W3159348040 endingPage "270" @default.
- W3159348040 startingPage "262" @default.
- W3159348040 abstract "Optimal transport (OT) distances are increasingly used as loss functions for statistical inference, notably in the learning of generative models or supervised learning. Yet, the behavior of minimum Wasserstein estimators is poorly understood, notably in high-dimensional regimes or under model misspecification. In this work we adopt the viewpoint of projection robust (PR) OT, which seeks to maximize the OT cost between two measures by choosing a $k$-dimensional subspace onto which they can be projected. Our first contribution is to establish several fundamental statistical properties of PR Wasserstein distances, complementing and improving previous literature that has been restricted to one-dimensional and well-specified cases. Next, we propose the integral PR Wasserstein (IPRW) distance as an alternative to the PRW distance, by averaging rather than optimizing on subspaces. Our complexity bounds can help explain why both PRW and IPRW distances outperform Wasserstein distances empirically in high-dimensional inference tasks. Finally, we consider parametric inference using the PRW distance. We provide an asymptotic guarantee of two types of minimum PRW estimators and formulate a central limit theorem for max-sliced Wasserstein estimator under model misspecification. To enable our analysis on PRW with projection dimension larger than one, we devise a novel combination of variational analysis and statistical theory." @default.
- W3159348040 created "2021-05-10" @default.
- W3159348040 creator A5002874750 @default.
- W3159348040 creator A5017263427 @default.
- W3159348040 creator A5049812527 @default.
- W3159348040 creator A5057412057 @default.
- W3159348040 creator A5083720030 @default.
- W3159348040 date "2021-03-18" @default.
- W3159348040 modified "2023-09-23" @default.
- W3159348040 title "On Projection Robust Optimal Transport: Sample Complexity and Model Misspecification" @default.
- W3159348040 hasPublicationYear "2021" @default.
- W3159348040 type Work @default.
- W3159348040 sameAs 3159348040 @default.
- W3159348040 citedByCount "12" @default.
- W3159348040 countsByYear W31593480402020 @default.
- W3159348040 countsByYear W31593480402021 @default.
- W3159348040 crossrefType "proceedings-article" @default.
- W3159348040 hasAuthorship W3159348040A5002874750 @default.
- W3159348040 hasAuthorship W3159348040A5017263427 @default.
- W3159348040 hasAuthorship W3159348040A5049812527 @default.
- W3159348040 hasAuthorship W3159348040A5057412057 @default.
- W3159348040 hasAuthorship W3159348040A5083720030 @default.
- W3159348040 hasConcept C105795698 @default.
- W3159348040 hasConcept C11413529 @default.
- W3159348040 hasConcept C117251300 @default.
- W3159348040 hasConcept C12362212 @default.
- W3159348040 hasConcept C126255220 @default.
- W3159348040 hasConcept C134261354 @default.
- W3159348040 hasConcept C154945302 @default.
- W3159348040 hasConcept C185429906 @default.
- W3159348040 hasConcept C202444582 @default.
- W3159348040 hasConcept C2524010 @default.
- W3159348040 hasConcept C2776214188 @default.
- W3159348040 hasConcept C28826006 @default.
- W3159348040 hasConcept C32834561 @default.
- W3159348040 hasConcept C33676613 @default.
- W3159348040 hasConcept C33923547 @default.
- W3159348040 hasConcept C41008148 @default.
- W3159348040 hasConcept C57493831 @default.
- W3159348040 hasConceptScore W3159348040C105795698 @default.
- W3159348040 hasConceptScore W3159348040C11413529 @default.
- W3159348040 hasConceptScore W3159348040C117251300 @default.
- W3159348040 hasConceptScore W3159348040C12362212 @default.
- W3159348040 hasConceptScore W3159348040C126255220 @default.
- W3159348040 hasConceptScore W3159348040C134261354 @default.
- W3159348040 hasConceptScore W3159348040C154945302 @default.
- W3159348040 hasConceptScore W3159348040C185429906 @default.
- W3159348040 hasConceptScore W3159348040C202444582 @default.
- W3159348040 hasConceptScore W3159348040C2524010 @default.
- W3159348040 hasConceptScore W3159348040C2776214188 @default.
- W3159348040 hasConceptScore W3159348040C28826006 @default.
- W3159348040 hasConceptScore W3159348040C32834561 @default.
- W3159348040 hasConceptScore W3159348040C33676613 @default.
- W3159348040 hasConceptScore W3159348040C33923547 @default.
- W3159348040 hasConceptScore W3159348040C41008148 @default.
- W3159348040 hasConceptScore W3159348040C57493831 @default.
- W3159348040 hasLocation W31593480401 @default.
- W3159348040 hasOpenAccess W3159348040 @default.
- W3159348040 hasPrimaryLocation W31593480401 @default.
- W3159348040 hasRelatedWork W1594039573 @default.
- W3159348040 hasRelatedWork W1639961155 @default.
- W3159348040 hasRelatedWork W2019106840 @default.
- W3159348040 hasRelatedWork W2158131535 @default.
- W3159348040 hasRelatedWork W2345894225 @default.
- W3159348040 hasRelatedWork W2739748921 @default.
- W3159348040 hasRelatedWork W2895248218 @default.
- W3159348040 hasRelatedWork W2914003818 @default.
- W3159348040 hasRelatedWork W2962829336 @default.
- W3159348040 hasRelatedWork W2963506485 @default.
- W3159348040 hasRelatedWork W2963897698 @default.
- W3159348040 hasRelatedWork W2964308809 @default.
- W3159348040 hasRelatedWork W2970112944 @default.
- W3159348040 hasRelatedWork W2979557588 @default.
- W3159348040 hasRelatedWork W2980104607 @default.
- W3159348040 hasRelatedWork W3003876312 @default.
- W3159348040 hasRelatedWork W3036776824 @default.
- W3159348040 hasRelatedWork W3104419752 @default.
- W3159348040 hasRelatedWork W3178171555 @default.
- W3159348040 hasRelatedWork W3187337999 @default.
- W3159348040 isParatext "false" @default.
- W3159348040 isRetracted "false" @default.
- W3159348040 magId "3159348040" @default.
- W3159348040 workType "article" @default.