Risk-sensitive generalizations of minimum variance estimation and control
This paper presents the risk-sensitive nonlinear stochastic filtering problem for general finite dimensional signal models, including discrete state hidden Markov models. The authors define a risk-sensitive filter as minimizing the expectation of the exponential of a quadratic error with risk-sensitive parameter and obtain the following results using arguments on the information state filter. 1. For Gaussian models, the risk-sensitive filter is identical to the Kalman filter. 2. As the risk-sensitive parameter approaches 0, the risk-sensitive filter tends to a risk-neutral filter. 3. In the small noise limit, the problem is related to a deterministic differential game, namely an \(H_\infty\) problem. 4. For discrete-time signal models, a risk-sensitive generalization of minimum variance control is given.
- On asymptotic stability of continuous-time risk-sensitive filters with respect to initial conditions
- Some partially observed multi-agent linear exponential quadratic stochastic differential games
- Risk-sensitive filtering, prediction and smoothing for discrete-time singular systems
- Generalised risk-sensitive control with full and partial state observation
- Risk sensitive identification of linear stochastic systems
- Event-triggered smoothing for hidden Markov models: risk-sensitive and MMSE results
- Deterministic and stochastic approaches to nonlinear filtering
- A contraction analysis of the convergence of risk-sensitive filters
- scientific article; zbMATH DE number 3948142 (Why is no real title available?)
- Risk-sensitive estimation and a differential game
- Risk-sensitive filtering and smoothing via reference probability methods
- scientific article; zbMATH DE number 1140227 (Why is no real title available?)
- scientific article; zbMATH DE number 2015385 (Why is no real title available?)
- scientific article; zbMATH DE number 1474998 (Why is no real title available?)
- scientific article; zbMATH DE number 1507332 (Why is no real title available?)
- New finite-dimensional risk-sensitive filters: small noise limits
- scientific article; zbMATH DE number 935562 (Why is no real title available?)
- Robust limits of risk sensitive nonlinear filters
- Adaptive risk-sensitive filter for Markovian jump linear systems
- Risk-sensitive filtering and smoothing for hidden Markov models
- Linear quadratic control with risk constraints
- Risk sensitive and LEG filtering problems are not equivalent
This page was built for publication: Risk-sensitive generalizations of minimum variance estimation and control
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4386557)
