Matches in SemOpenAlex for { <https://semopenalex.org/work/W4320342859> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W4320342859 abstract "Transport maps can ease the sampling of distributions with non-trivial geometries by transforming them into distributions that are easier to handle. The potential of this approach has risen with the development of Normalizing Flows (NF) which are maps parameterized with deep neural networks trained to push a reference distribution towards a target. NF-enhanced samplers recently proposed blend (Markov chain) Monte Carlo methods with either (i) proposal draws from the flow or (ii) a flow-based reparametrization. In both cases, the quality of the learned transport conditions performance. The present work clarifies for the first time the relative strengths and weaknesses of these two approaches. Our study concludes that multimodal targets can be reliably handled with flow-based proposals up to moderately high dimensions. In contrast, methods relying on reparametrization struggle with multimodality but are more robust otherwise in high-dimensional settings and under poor training. To further illustrate the influence of target-proposal adequacy, we also derive a new quantitative bound for the mixing time of the Independent Metropolis-Hastings sampler." @default.
- W4320342859 created "2023-02-13" @default.
- W4320342859 creator A5000295817 @default.
- W4320342859 creator A5001522727 @default.
- W4320342859 creator A5036096413 @default.
- W4320342859 creator A5048873755 @default.
- W4320342859 date "2023-02-09" @default.
- W4320342859 modified "2023-09-24" @default.
- W4320342859 title "On Sampling with Approximate Transport Maps" @default.
- W4320342859 doi "https://doi.org/10.48550/arxiv.2302.04763" @default.
- W4320342859 hasPublicationYear "2023" @default.
- W4320342859 type Work @default.
- W4320342859 citedByCount "0" @default.
- W4320342859 crossrefType "posted-content" @default.
- W4320342859 hasAuthorship W4320342859A5000295817 @default.
- W4320342859 hasAuthorship W4320342859A5001522727 @default.
- W4320342859 hasAuthorship W4320342859A5036096413 @default.
- W4320342859 hasAuthorship W4320342859A5048873755 @default.
- W4320342859 hasBestOaLocation W43203428591 @default.
- W4320342859 hasConcept C105795698 @default.
- W4320342859 hasConcept C106131492 @default.
- W4320342859 hasConcept C107673813 @default.
- W4320342859 hasConcept C110121322 @default.
- W4320342859 hasConcept C111350023 @default.
- W4320342859 hasConcept C111472728 @default.
- W4320342859 hasConcept C11413529 @default.
- W4320342859 hasConcept C119857082 @default.
- W4320342859 hasConcept C121332964 @default.
- W4320342859 hasConcept C134306372 @default.
- W4320342859 hasConcept C138885662 @default.
- W4320342859 hasConcept C140779682 @default.
- W4320342859 hasConcept C154945302 @default.
- W4320342859 hasConcept C165464430 @default.
- W4320342859 hasConcept C19499675 @default.
- W4320342859 hasConcept C2524010 @default.
- W4320342859 hasConcept C2776502983 @default.
- W4320342859 hasConcept C2779530757 @default.
- W4320342859 hasConcept C31972630 @default.
- W4320342859 hasConcept C33923547 @default.
- W4320342859 hasConcept C38349280 @default.
- W4320342859 hasConcept C41008148 @default.
- W4320342859 hasConcept C62520636 @default.
- W4320342859 hasConcept C63882131 @default.
- W4320342859 hasConcept C98763669 @default.
- W4320342859 hasConceptScore W4320342859C105795698 @default.
- W4320342859 hasConceptScore W4320342859C106131492 @default.
- W4320342859 hasConceptScore W4320342859C107673813 @default.
- W4320342859 hasConceptScore W4320342859C110121322 @default.
- W4320342859 hasConceptScore W4320342859C111350023 @default.
- W4320342859 hasConceptScore W4320342859C111472728 @default.
- W4320342859 hasConceptScore W4320342859C11413529 @default.
- W4320342859 hasConceptScore W4320342859C119857082 @default.
- W4320342859 hasConceptScore W4320342859C121332964 @default.
- W4320342859 hasConceptScore W4320342859C134306372 @default.
- W4320342859 hasConceptScore W4320342859C138885662 @default.
- W4320342859 hasConceptScore W4320342859C140779682 @default.
- W4320342859 hasConceptScore W4320342859C154945302 @default.
- W4320342859 hasConceptScore W4320342859C165464430 @default.
- W4320342859 hasConceptScore W4320342859C19499675 @default.
- W4320342859 hasConceptScore W4320342859C2524010 @default.
- W4320342859 hasConceptScore W4320342859C2776502983 @default.
- W4320342859 hasConceptScore W4320342859C2779530757 @default.
- W4320342859 hasConceptScore W4320342859C31972630 @default.
- W4320342859 hasConceptScore W4320342859C33923547 @default.
- W4320342859 hasConceptScore W4320342859C38349280 @default.
- W4320342859 hasConceptScore W4320342859C41008148 @default.
- W4320342859 hasConceptScore W4320342859C62520636 @default.
- W4320342859 hasConceptScore W4320342859C63882131 @default.
- W4320342859 hasConceptScore W4320342859C98763669 @default.
- W4320342859 hasLocation W43203428591 @default.
- W4320342859 hasOpenAccess W4320342859 @default.
- W4320342859 hasPrimaryLocation W43203428591 @default.
- W4320342859 hasRelatedWork W1593554773 @default.
- W4320342859 hasRelatedWork W1594831322 @default.
- W4320342859 hasRelatedWork W1928096970 @default.
- W4320342859 hasRelatedWork W1965870577 @default.
- W4320342859 hasRelatedWork W2031427063 @default.
- W4320342859 hasRelatedWork W2036896164 @default.
- W4320342859 hasRelatedWork W2086235234 @default.
- W4320342859 hasRelatedWork W241360418 @default.
- W4320342859 hasRelatedWork W4248228406 @default.
- W4320342859 hasRelatedWork W74710183 @default.
- W4320342859 isParatext "false" @default.
- W4320342859 isRetracted "false" @default.
- W4320342859 workType "article" @default.