We consider filtering in high-dimensional non-Gaussian state-space models with intractable transition kernels, nonlinear and possibly chaotic dynamics, and sparse observations in space and time. We propose a novel filtering methodology that harnesses transportation of measures, convex optimization, and ideas from probabilistic graphical models to yield robust ensemble approximations of the filtering distribution in high dimensions. Our approach can be understood as the natural generalization of the ensemble Kalman filter (EnKF) to nonlinear updates, using stochastic or deterministic couplings. The use of nonlinear updates can reduce the intrinsic bias of the EnKF at a marginal increase in computational cost. We avoid any form of importance sampling and introduce non-Gaussian localization approaches for dimension scalability. Our framework achieves state-of-the-art tracking performance on challenging configurations of the Lorenz-96 model in the chaotic regime.
{"title":"Coupling Techniques for Nonlinear Ensemble Filtering","authors":"Alessio Spantini, R. Baptista, Y. Marzouk","doi":"10.1137/20m1312204","DOIUrl":"https://doi.org/10.1137/20m1312204","url":null,"abstract":"We consider filtering in high-dimensional non-Gaussian state-space models with intractable transition kernels, nonlinear and possibly chaotic dynamics, and sparse observations in space and time. We propose a novel filtering methodology that harnesses transportation of measures, convex optimization, and ideas from probabilistic graphical models to yield robust ensemble approximations of the filtering distribution in high dimensions. Our approach can be understood as the natural generalization of the ensemble Kalman filter (EnKF) to nonlinear updates, using stochastic or deterministic couplings. The use of nonlinear updates can reduce the intrinsic bias of the EnKF at a marginal increase in computational cost. We avoid any form of importance sampling and introduce non-Gaussian localization approaches for dimension scalability. Our framework achieves state-of-the-art tracking performance on challenging configurations of the Lorenz-96 model in the chaotic regime.","PeriodicalId":49525,"journal":{"name":"SIAM Review","volume":" 19","pages":""},"PeriodicalIF":10.2,"publicationDate":"2019-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41252399","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1984-07-01DOI: 10.1007/978-1-4757-3240-5_56
J. Borwein, P. Borwein
{"title":"The Arithmetic-Geometric Mean and Fast Computation of Elementary Functions","authors":"J. Borwein, P. Borwein","doi":"10.1007/978-1-4757-3240-5_56","DOIUrl":"https://doi.org/10.1007/978-1-4757-3240-5_56","url":null,"abstract":"","PeriodicalId":49525,"journal":{"name":"SIAM Review","volume":"18 1","pages":"537-552"},"PeriodicalIF":10.2,"publicationDate":"1984-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"51028130","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1983-01-01DOI: 10.1007/978-3-642-60543-7_4
A. Harten, P. Lax, B. Leer
{"title":"On Upstream Differencing and Godunov-Type Schemes for Hyperbolic Conservation Laws","authors":"A. Harten, P. Lax, B. Leer","doi":"10.1007/978-3-642-60543-7_4","DOIUrl":"https://doi.org/10.1007/978-3-642-60543-7_4","url":null,"abstract":"","PeriodicalId":49525,"journal":{"name":"SIAM Review","volume":"68 1","pages":"53-79"},"PeriodicalIF":10.2,"publicationDate":"1983-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1007/978-3-642-60543-7_4","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"51110818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：