Approximate Bayes Linear Smoothers for Continuous Processes
نویسنده
چکیده
This paper presents two approximations for the problem of Bayes linear estimation of continuous vector-valued stochastic processes. The problem of interpolating time series is treated as a particular application. Expressions for errors of approximations are obtained. In particular, our general conditions for zero error lead to a characterization of a generalized ARC!) process.
منابع مشابه
Continuous-time and continuous-discrete-time unscented Rauch-Tung-Striebel smoothers
This article considers the application of the unscented transformation to approximate fixed-interval optimal smoothing of continuous-time non-linear stochastic systems. The proposed methodology can be applied to systems, where the dynamics can be modeled with non-linear stochastic differential equations and the noise corrupted measurements are obtained continuously or at discrete times. The smo...
متن کاملSmoothers for Discontinuous Signals
First we explain the interplay between robust loss functions non linear lters and Bayes smoothers for edge preserving image recon struction Then we prove the surprising fact that maximum posterior smoothers are nonlinear lters A generalized Potts prior for segmen tation and piecewise smoothing of noisy signals and images is adopted For one dimensional signals an exact solution for the maximum p...
متن کاملRobust continuous-time smoothers without two-sided stochastic integrals
We consider the problem of fixed-interval smoothing of a continuous-time partially observed nonlinear stochastic dynamical system. Existing results for such smoothers require the use of two-sided stochastic calculus. The main contribution of this paper is to present a robust formulation of the smoothing equations. Under this robust formulation, the smoothing equations are nonstochastic paraboli...
متن کاملCombining Multilevel Preconditioners with Domain Decomposition for Solving Linear Systems
Algebraic multilevel techniques for defining preconditioners have proven to bevery effective when used with Krylov methods. In most cases, they can solvelinear systems arising from the dicretization of PDEs with a number of floatingpoint operations proportional to the number of unknowns. To be efficient onparallel computers these multilevel techniques must use parallel smoothers...
متن کاملOptimal Bayesian Online Learning
In a Bayesian approach to online learning a simple paramet-ric approximate posterior over rules is updated in each online learning step. Predictions on new data are derived from averages over this posterior. This should be compared to the Bayes optimal batch (or ooine) approach for which the posterior is calculated from the prior and the likelihood of the whole training set. We suggest that min...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2011