FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2593254544> ?p ?o ?g. }

• W2593254544 abstract "It is well known in many settings that reversible Langevin diffusions in confining potentials converge to equilibrium exponentially fast. Adding irreversible perturbations to the drift of a Langevin diffusion that maintain the same invariant measure accelerates its convergence to stationarity. Many existing works thus advocate the use of such non-reversible dynamics for sampling. When implementing Markov Chain Monte Carlo algorithms (MCMC) using time discretisations of such Stochastic Differential Equations (SDEs), one can append the discretization with the usual Metropolis-Hastings accept-reject step and this is often done in practice because the accept--reject step eliminates bias. On the other hand, such a step makes the resulting chain reversible. It is not known whether adding the accept-reject step preserves the faster mixing properties of the non-reversible dynamics. In this paper, we address this gap between theory and practice by analyzing the optimal scaling of MCMC algorithms constructed from proposal moves that are time-step Euler discretisations of an irreversible SDE, for high dimensional Gaussian target measures. We call the resulting algorithm the imala, in comparison to the classical MALA algorithm (here {em ip} is for irreversible proposal). In order to quantify how the cost of the algorithm scales with the dimension $N$, we prove invariance principles for the appropriately rescaled chain. In contrast to the usual MALA algorithm, we show that there could be two regimes asymptotically: (i) a diffusive regime, as in the MALA algorithm and (ii) a ``fluid regime where the limit is an ordinary differential equation. We provide concrete examples where the limit is a diffusion, as in the standard MALA, but with provably higher limiting acceptance probabilities. Numerical results are also given corroborating the theory." @default.
• W2593254544 created "2017-03-16" @default.
• W2593254544 creator A5002796692 @default.
• W2593254544 creator A5069207970 @default.
• W2593254544 creator A5073501417 @default.
• W2593254544 date "2017-02-06" @default.
• W2593254544 modified "2023-09-27" @default.
• W2593254544 title "Optimal Scaling of the MALA algorithm with Irreversible Proposals for Gaussian targets" @default.
• W2593254544 cites W1599068951 @default.
• W2593254544 cites W1872439012 @default.
• W2593254544 cites W1922512035 @default.
• W2593254544 cites W1975742473 @default.
• W2593254544 cites W1983452151 @default.
• W2593254544 cites W2006367902 @default.
• W2593254544 cites W2008703230 @default.
• W2593254544 cites W2029164135 @default.
• W2593254544 cites W2035081607 @default.
• W2593254544 cites W2057291289 @default.
• W2593254544 cites W2090251478 @default.
• W2593254544 cites W2118847508 @default.
• W2593254544 cites W2153811005 @default.
• W2593254544 cites W2253619190 @default.
• W2593254544 cites W2467132312 @default.
• W2593254544 cites W2497694760 @default.
• W2593254544 cites W2514112094 @default.
• W2593254544 cites W2573128940 @default.
• W2593254544 cites W2581130666 @default.
• W2593254544 cites W2962732346 @default.
• W2593254544 cites W2964086105 @default.
• W2593254544 cites W2999494274 @default.
• W2593254544 cites W3037124231 @default.
• W2593254544 cites W3101896223 @default.
• W2593254544 cites W3102325779 @default.
• W2593254544 hasPublicationYear "2017" @default.
• W2593254544 type Work @default.
• W2593254544 sameAs 2593254544 @default.
• W2593254544 citedByCount "2" @default.
• W2593254544 countsByYear W25932545442017 @default.
• W2593254544 countsByYear W25932545442019 @default.
• W2593254544 crossrefType "posted-content" @default.
• W2593254544 hasAuthorship W2593254544A5002796692 @default.
• W2593254544 hasAuthorship W2593254544A5069207970 @default.
• W2593254544 hasAuthorship W2593254544A5073501417 @default.
• W2593254544 hasConcept C105795698 @default.
• W2593254544 hasConcept C111350023 @default.
• W2593254544 hasConcept C11413529 @default.
• W2593254544 hasConcept C121332964 @default.
• W2593254544 hasConcept C121864883 @default.
• W2593254544 hasConcept C134306372 @default.
• W2593254544 hasConcept C162324750 @default.
• W2593254544 hasConcept C163716315 @default.
• W2593254544 hasConcept C19499675 @default.
• W2593254544 hasConcept C204693719 @default.
• W2593254544 hasConcept C2524010 @default.
• W2593254544 hasConcept C2777303404 @default.
• W2593254544 hasConcept C2780004032 @default.
• W2593254544 hasConcept C28826006 @default.
• W2593254544 hasConcept C33923547 @default.
• W2593254544 hasConcept C41008148 @default.
• W2593254544 hasConcept C50522688 @default.
• W2593254544 hasConcept C51544822 @default.
• W2593254544 hasConcept C51955184 @default.
• W2593254544 hasConcept C62520636 @default.
• W2593254544 hasConcept C73000952 @default.
• W2593254544 hasConcept C78045399 @default.
• W2593254544 hasConcept C98763669 @default.
• W2593254544 hasConcept C99844830 @default.
• W2593254544 hasConceptScore W2593254544C105795698 @default.
• W2593254544 hasConceptScore W2593254544C111350023 @default.
• W2593254544 hasConceptScore W2593254544C11413529 @default.
• W2593254544 hasConceptScore W2593254544C121332964 @default.
• W2593254544 hasConceptScore W2593254544C121864883 @default.
• W2593254544 hasConceptScore W2593254544C134306372 @default.
• W2593254544 hasConceptScore W2593254544C162324750 @default.
• W2593254544 hasConceptScore W2593254544C163716315 @default.
• W2593254544 hasConceptScore W2593254544C19499675 @default.
• W2593254544 hasConceptScore W2593254544C204693719 @default.
• W2593254544 hasConceptScore W2593254544C2524010 @default.
• W2593254544 hasConceptScore W2593254544C2777303404 @default.
• W2593254544 hasConceptScore W2593254544C2780004032 @default.
• W2593254544 hasConceptScore W2593254544C28826006 @default.
• W2593254544 hasConceptScore W2593254544C33923547 @default.
• W2593254544 hasConceptScore W2593254544C41008148 @default.
• W2593254544 hasConceptScore W2593254544C50522688 @default.
• W2593254544 hasConceptScore W2593254544C51544822 @default.
• W2593254544 hasConceptScore W2593254544C51955184 @default.
• W2593254544 hasConceptScore W2593254544C62520636 @default.
• W2593254544 hasConceptScore W2593254544C73000952 @default.
• W2593254544 hasConceptScore W2593254544C78045399 @default.
• W2593254544 hasConceptScore W2593254544C98763669 @default.
• W2593254544 hasConceptScore W2593254544C99844830 @default.
• W2593254544 hasLocation W25932545441 @default.
• W2593254544 hasOpenAccess W2593254544 @default.
• W2593254544 hasPrimaryLocation W25932545441 @default.
• W2593254544 hasRelatedWork W1575284787 @default.
• W2593254544 hasRelatedWork W170092604 @default.
• W2593254544 hasRelatedWork W1984720857 @default.
• W2593254544 hasRelatedWork W2090231716 @default.
• W2593254544 hasRelatedWork W2776466161 @default.
• W2593254544 hasRelatedWork W2940336419 @default.

• next