Symmetrically Processed Splitting Integrators for Enhanced Hamiltonian Monte Carlo Sampling
| Autor: | Fernando Casas, Jesús María Sanz-Serna, Sergio Blanes, M. P. Calvo |
|---|---|
| Přispěvatelé: | Blanes, Sergio |
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Hamiltonian Monte Carlo method
Applied Mathematics Monte Carlo method Sampling (statistics) Numerical Analysis (math.NA) Processing Splitting integrators Hybrid Monte Carlo Computational Mathematics Integrator FOS: Mathematics Applied mathematics Mathematics - Numerical Analysis MATEMATICA APLICADA Hamiltonian (control theory) Computational budget Mathematics |
| Zdroj: | RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia instname Repositori Universitat Jaume I Universitat Jaume I |
| ISSN: | 2019-1049 |
| DOI: | 10.1137/20M137940X30 |
| Popis: | [EN] We construct integrators to be used in Hamiltonian (or Hybrid) Monte Carlo sampling. The new integrators are easily implementable and, for a given computational budget, may deliver five times as many accepted proposals as standard leapfrog/Verlet without impairing in any way the quality of the samples. They are based on a suitable modification of the processing technique first introduced by Butcher. The idea of modified processing may also be useful for other purposes, like the construction of high-order splitting integrators with positive coefficients. The first, third, and fourth authors were supported by project PID2019-104927GB-C21 (AEI/FEDER, UE) . The second author was supported by projects PID2019-104927GB-C22 (GNI-QUAMC) , (AEI/FEDER, UE) VA105G18, and VA169P20 (Junta de Castilla y Leon, ES) co-financed by FEDER funds. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|