Markovian stochastic approximation with expanding projections
Stochastic approximation is a framework unifying many random iterative algorithms occurring in a diverse range of applications. The stability of the process is often difficult to verify in practical applications and the process may even be unstable without additional stabilisation techniques. We stu...
Main Authors: | , |
---|---|
Format: | Journal article |
Language: | English |
Published: |
International Statistical Institute
2014
|
_version_ | 1797054444692570112 |
---|---|
author | Andrieu, C Vihola, M |
author_facet | Andrieu, C Vihola, M |
author_sort | Andrieu, C |
collection | OXFORD |
description | Stochastic approximation is a framework unifying many random iterative algorithms occurring in a diverse range of applications. The stability of the process is often difficult to verify in practical applications and the process may even be unstable without additional stabilisation techniques. We study a stochastic approximation procedure with expanding projections similar to Andradóttir [Oper. Res. 43 (1995) 1037-1048]. We focus on Markovian noise and show the stability and convergence under general conditions. Our framework also incorporates the possibility to use a random step size sequence, which allows us to consider settings with a non-smooth family of Markov kernels. We apply the theory to stochastic approximation expectation maximisation with particle independent Metropolis-Hastings sampling. © 2014 ISI/BS. |
first_indexed | 2024-03-06T18:57:23Z |
format | Journal article |
id | oxford-uuid:124f47c7-94ac-4d6f-9435-abc6db1628ba |
institution | University of Oxford |
language | English |
last_indexed | 2024-03-06T18:57:23Z |
publishDate | 2014 |
publisher | International Statistical Institute |
record_format | dspace |
spelling | oxford-uuid:124f47c7-94ac-4d6f-9435-abc6db1628ba2022-03-26T10:07:12ZMarkovian stochastic approximation with expanding projectionsJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:124f47c7-94ac-4d6f-9435-abc6db1628baEnglishSymplectic Elements at OxfordInternational Statistical Institute2014Andrieu, CVihola, MStochastic approximation is a framework unifying many random iterative algorithms occurring in a diverse range of applications. The stability of the process is often difficult to verify in practical applications and the process may even be unstable without additional stabilisation techniques. We study a stochastic approximation procedure with expanding projections similar to Andradóttir [Oper. Res. 43 (1995) 1037-1048]. We focus on Markovian noise and show the stability and convergence under general conditions. Our framework also incorporates the possibility to use a random step size sequence, which allows us to consider settings with a non-smooth family of Markov kernels. We apply the theory to stochastic approximation expectation maximisation with particle independent Metropolis-Hastings sampling. © 2014 ISI/BS. |
spellingShingle | Andrieu, C Vihola, M Markovian stochastic approximation with expanding projections |
title | Markovian stochastic approximation with expanding projections |
title_full | Markovian stochastic approximation with expanding projections |
title_fullStr | Markovian stochastic approximation with expanding projections |
title_full_unstemmed | Markovian stochastic approximation with expanding projections |
title_short | Markovian stochastic approximation with expanding projections |
title_sort | markovian stochastic approximation with expanding projections |
work_keys_str_mv | AT andrieuc markovianstochasticapproximationwithexpandingprojections AT viholam markovianstochasticapproximationwithexpandingprojections |