An Optimal Transport Formulation of Bayes' Law for Nonlinear Filtering Algorithms

被引:13
作者
Taghvaei, Amirhossein [1 ]
Hosseini, Bamdad [2 ]
机构
[1] Univ Washington, Dept Aeronaut & Astronaut, Seattle, WA 98195 USA
[2] Univ Washington, Dept Appl Math, Seattle, WA 98195 USA
来源
2022 IEEE 61ST CONFERENCE ON DECISION AND CONTROL (CDC) | 2022年
关键词
D O I
10.1109/CDC51059.2022.9992776
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This paper presents a variational representation of the Bayes' law using optimal transportation theory. The variational representation is in terms of the optimal transportation between the joint distribution of the (state, observation) and their independent coupling. By imposing certain structure on the transport map, the solution to the variational problem is used to construct a Brenier-type map that transports the prior distribution to the posterior distribution for any value of the observation signal. The new formulation is used to derive the optimal transport form of the Ensemble Kalman filter (EnKF) for the discrete-time filtering problem and propose a novel extension of EnKF to the non-Gaussian setting utilizing input convex neural networks. Finally, the proposed methodology is used to derive the optimal transport form of the feedback particle filler (FPF) in the continuous-time limit, which constitutes its first variational construction without explicitly using the nonlinear filtering equation or Bayes' law.
引用
收藏
页码:6608 / 6613
页数:6
相关论文
共 40 条
[1]  
Amos B, 2017, PR MACH LEARN RES, V70
[2]  
[Anonymous], 2010, SPIE DEFENSE SECURIT
[3]  
Bengtsson T., 2008, IMS Collections: Probability and Statistics: Essays in Honor of David A. Freedman, P316, DOI DOI 10.1214/193940307000000518
[4]   An ensemble Kalman-Bucy filter for continuous data assimilation [J].
Bergemann, Kay ;
Reich, Sebastian .
METEOROLOGISCHE ZEITSCHRIFT, 2012, 21 (03) :213-219
[5]  
Beskos A, 2014, ADV APPL PROBAB, V46, P279
[6]  
BRENIER Y, 1987, CR ACAD SCI I-MATH, V305, P805
[7]   VECTOR QUANTILE REGRESSION: AN OPTIMAL TRANSPORT APPROACH [J].
Carlier, Guillaume ;
Chernozhukov, Victor ;
Galichon, Alfred .
ANNALS OF STATISTICS, 2016, 44 (03) :1165-1192
[8]  
Chen Y., 2018, Optimal Control via Neural Networks: A Convex Approach
[9]  
Cheng Yuan, 2013, ARXIV13116300
[10]   Approximate McKean-Vlasov representations for a class of SPDEs [J].
Crisan, Dan ;
Xiong, Jie .
STOCHASTICS-AN INTERNATIONAL JOURNAL OF PROBABILITY AND STOCHASTIC PROCESSES, 2010, 82 (01) :53-68