No Cover Image

Technical Report 381 views

Variational Bayesian inference of hidden stochastic processes with unknown parameters

Pavel Loskot Orcid Logo, Komlan Atitey

Swansea University Authors: Pavel Loskot Orcid Logo, Komlan Atitey

Abstract

Estimating hidden processes from non-linear noisy observations is particularly difficult when the parameters of these processes are not known. This paper adopts a machine learning approach to devise variational Bayesian inference for such scenarios. In particular, a random process generated by the a...

Full description

Published:
Online Access: https://arxiv.org/abs/1911.00757
URI: https://cronfa.swan.ac.uk/Record/cronfa52652
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2019-11-05T19:13:45Z
last_indexed 2023-04-18T03:05:53Z
id cronfa52652
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2023-04-17T10:52:22.2957978</datestamp><bib-version>v2</bib-version><id>52652</id><entry>2019-11-05</entry><title>Variational Bayesian inference of hidden stochastic processes with unknown parameters</title><swanseaauthors><author><sid>bc7cba9ef306864239b9348c3aea4c3e</sid><ORCID>0000-0002-2773-2186</ORCID><firstname>Pavel</firstname><surname>Loskot</surname><name>Pavel Loskot</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>556d8b85e41c10e686245848e3f080b6</sid><firstname>Komlan</firstname><surname>Atitey</surname><name>Komlan Atitey</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2019-11-05</date><deptcode>EEN</deptcode><abstract>Estimating hidden processes from non-linear noisy observations is particularly difficult when the parameters of these processes are not known. This paper adopts a machine learning approach to devise variational Bayesian inference for such scenarios. In particular, a random process generated by the autoregressive moving average (ARMA) linear model is inferred from non-linearity noise observations. The posterior distribution of hidden states are approximated by a set of weighted particles generated by the sequential Monte carlo (SMC) algorithm involving sampling with importance sampling resampling (SISR). Numerical efficiency and estimation accuracy of the proposed inference method are evaluated by computer simulations. Furthermore, the proposed inference method is demonstrated on a practical problem of estimating the missing values in the gene expression time series assuming vector autoregressive (VAR) data model.</abstract><type>Technical Report</type><journal/><volume/><journalNumber/><paginationStart/><paginationEnd/><publisher/><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic/><keywords/><publishedDay>0</publishedDay><publishedMonth>0</publishedMonth><publishedYear>0</publishedYear><publishedDate>0001-01-01</publishedDate><doi/><url>https://arxiv.org/abs/1911.00757</url><notes/><college>COLLEGE NANME</college><department>Engineering</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>EEN</DepartmentCode><institution>Swansea University</institution><apcterm/><funders/><projectreference/><lastEdited>2023-04-17T10:52:22.2957978</lastEdited><Created>2019-11-05T15:44:50.8154200</Created><path><level id="1"/><level id="2"/></path><authors><author><firstname>Pavel</firstname><surname>Loskot</surname><orcid>0000-0002-2773-2186</orcid><order>1</order></author><author><firstname>Komlan</firstname><surname>Atitey</surname><order>2</order></author></authors><documents/><OutputDurs/></rfc1807>
spelling 2023-04-17T10:52:22.2957978 v2 52652 2019-11-05 Variational Bayesian inference of hidden stochastic processes with unknown parameters bc7cba9ef306864239b9348c3aea4c3e 0000-0002-2773-2186 Pavel Loskot Pavel Loskot true false 556d8b85e41c10e686245848e3f080b6 Komlan Atitey Komlan Atitey true false 2019-11-05 EEN Estimating hidden processes from non-linear noisy observations is particularly difficult when the parameters of these processes are not known. This paper adopts a machine learning approach to devise variational Bayesian inference for such scenarios. In particular, a random process generated by the autoregressive moving average (ARMA) linear model is inferred from non-linearity noise observations. The posterior distribution of hidden states are approximated by a set of weighted particles generated by the sequential Monte carlo (SMC) algorithm involving sampling with importance sampling resampling (SISR). Numerical efficiency and estimation accuracy of the proposed inference method are evaluated by computer simulations. Furthermore, the proposed inference method is demonstrated on a practical problem of estimating the missing values in the gene expression time series assuming vector autoregressive (VAR) data model. Technical Report 0 0 0 0001-01-01 https://arxiv.org/abs/1911.00757 COLLEGE NANME Engineering COLLEGE CODE EEN Swansea University 2023-04-17T10:52:22.2957978 2019-11-05T15:44:50.8154200 Pavel Loskot 0000-0002-2773-2186 1 Komlan Atitey 2
title Variational Bayesian inference of hidden stochastic processes with unknown parameters
spellingShingle Variational Bayesian inference of hidden stochastic processes with unknown parameters
Pavel Loskot
Komlan Atitey
title_short Variational Bayesian inference of hidden stochastic processes with unknown parameters
title_full Variational Bayesian inference of hidden stochastic processes with unknown parameters
title_fullStr Variational Bayesian inference of hidden stochastic processes with unknown parameters
title_full_unstemmed Variational Bayesian inference of hidden stochastic processes with unknown parameters
title_sort Variational Bayesian inference of hidden stochastic processes with unknown parameters
author_id_str_mv bc7cba9ef306864239b9348c3aea4c3e
556d8b85e41c10e686245848e3f080b6
author_id_fullname_str_mv bc7cba9ef306864239b9348c3aea4c3e_***_Pavel Loskot
556d8b85e41c10e686245848e3f080b6_***_Komlan Atitey
author Pavel Loskot
Komlan Atitey
author2 Pavel Loskot
Komlan Atitey
format Technical Report
institution Swansea University
url https://arxiv.org/abs/1911.00757
document_store_str 0
active_str 0
description Estimating hidden processes from non-linear noisy observations is particularly difficult when the parameters of these processes are not known. This paper adopts a machine learning approach to devise variational Bayesian inference for such scenarios. In particular, a random process generated by the autoregressive moving average (ARMA) linear model is inferred from non-linearity noise observations. The posterior distribution of hidden states are approximated by a set of weighted particles generated by the sequential Monte carlo (SMC) algorithm involving sampling with importance sampling resampling (SISR). Numerical efficiency and estimation accuracy of the proposed inference method are evaluated by computer simulations. Furthermore, the proposed inference method is demonstrated on a practical problem of estimating the missing values in the gene expression time series assuming vector autoregressive (VAR) data model.
published_date 0001-01-01T04:05:08Z
_version_ 1763753390197702656
score 11.012678

Similar Items