{"title":"Error analysis of a discretization for stochastic linear quadratic control problems governed by SDEs","authors":"Yanqing Wang","doi":"10.1093/imamci/dnab031","DOIUrl":null,"url":null,"abstract":"In this work, a time-implicit discretization for stochastic linear quadratic problems subject to stochastic differential equations with control-dependence noises is proposed, and the convergence rate of this discretization is proved. Compared to the existing results, the control variables are stochastic processes and can be contained in systems’ diffusion term. Based on this discretization, a gradient descent algorithm and its convergence rate are presented. Finally, a numerical example is provided to support the theoretical finding.","PeriodicalId":56128,"journal":{"name":"IMA Journal of Mathematical Control and Information","volume":"38 4","pages":"1148-1173"},"PeriodicalIF":1.6000,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IMA Journal of Mathematical Control and Information","FirstCategoryId":"94","ListUrlMain":"https://ieeexplore.ieee.org/document/9646589/","RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"AUTOMATION & CONTROL SYSTEMS","Score":null,"Total":0}
引用次数: 3
Abstract
In this work, a time-implicit discretization for stochastic linear quadratic problems subject to stochastic differential equations with control-dependence noises is proposed, and the convergence rate of this discretization is proved. Compared to the existing results, the control variables are stochastic processes and can be contained in systems’ diffusion term. Based on this discretization, a gradient descent algorithm and its convergence rate are presented. Finally, a numerical example is provided to support the theoretical finding.
期刊介绍:
The Journal is to provide an outlet for papers which are original and of high quality in mathematical control theory, systems theory, and applied information sciences. Short papers and mathematical correspondence or technical notes will be welcome, although the primary function of the journal is to publish papers of substantial length and coverage. The emphasis will be upon relevance, originality and clarify of presentation, although timeliness may well be an important feature in acceptable papers. Speculative papers that suggest new avenues for research or potential solutions to unsolved problems of control and information theory will be particularly welcome. Specific application papers will not normally be within the remit of the journal. Applications that illustrate techniques or theories will be acceptable. A prime function of the journal is to encourage the interplay between control and information theory and other mathematical sciences.
All submitted papers will be judged on their merits by at least two referees and a full paper report will be available to the intending authors. Submitted articles will in general be published in an issue within six months of submission. Papers should not have previously published, nor should they be undes consideration for publication in another journal. This Journal takes publication ethics very seriously. If misconduct is found or suspected after the manuscript is published, the journal will investigate the matter and this may result in the article subsequently being retracted.