Variance Decay Property for Filter Stability

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Jin Won, Mehta, Prashant G.
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913405076504576
author Kim, Jin Won
Mehta, Prashant G.
author_facet Kim, Jin Won
Mehta, Prashant G.
contents This paper is concerned with the problem of nonlinear (stochastic) filter stability of a hidden Markov model (HMM) with white noise observations. A contribution is the variance decay property which is used to conclude filter stability. For this purpose, a new notion of the Poincaré inequality (PI) is introduced for the nonlinear filter. PI is related to both the ergodicity of the Markov process as well as the observability of the HMM. The proofs are based upon a recently discovered minimum variance duality which is used to transform the nonlinear filtering problem into a stochastic optimal control problem for a backward stochastic differential equation (BSDE).
format Preprint
id arxiv_https___arxiv_org_abs_2305_12850
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Variance Decay Property for Filter Stability
Kim, Jin Won
Mehta, Prashant G.
Optimization and Control
Probability
This paper is concerned with the problem of nonlinear (stochastic) filter stability of a hidden Markov model (HMM) with white noise observations. A contribution is the variance decay property which is used to conclude filter stability. For this purpose, a new notion of the Poincaré inequality (PI) is introduced for the nonlinear filter. PI is related to both the ergodicity of the Markov process as well as the observability of the HMM. The proofs are based upon a recently discovered minimum variance duality which is used to transform the nonlinear filtering problem into a stochastic optimal control problem for a backward stochastic differential equation (BSDE).
title Variance Decay Property for Filter Stability
topic Optimization and Control
Probability
url https://arxiv.org/abs/2305.12850