Analysis of the feedback particle filter with diffusion map based approximation of the gain

  title={Analysis of the feedback particle filter with diffusion map based approximation of the gain},
  author={Sahani Pathiraja and Wilhelm Stannat},
  journal={Foundations of Data Science},
<p style='text-indent:20px;'>Control-type particle filters have been receiving increasing attention over the last decade as a means of obtaining sample based approximations to the sequential Bayesian filtering problem in the nonlinear setting. Here we analyse one such type, namely the feedback particle filter and a recently proposed approximation of the associated gain function based on diffusion maps. The key purpose is to provide analytic insights on the form of the approximate gain, which… 

Rough McKean-Vlasov dynamics for robust ensemble Kalman filtering

A McKean-Vlasov equation that contains the data stream as a common driving rough path is studied, establishing propagation of chaos for the associated interacting particle system, which is suggestive of a numerical scheme that can be viewed as an extension of the ensemble Kalman filter to a rough-path framework.



A mean-field control-oriented approach to particle filtering

A new formulation of the particle filter for non linear filtering is presented, based on concepts from optimal control, and from the mean-field game theory framework of Huang et. al.. The optimal

Poisson's equation in nonlinear filtering

A time-stepping procedure is introduced, consisting of successive minimization problems in the space of probability densities, and the weak form of the nonlinear filter is derived via analysis of the first-order optimality conditions for these problems.

Feedback Particle Filter

The feedback particle filter introduced in this paper is a new approach to approximate nonlinear filtering, motivated by techniques from mean-field game theory, and numerical algorithms are introduced and implemented in two general examples, and a neuroscience application involving coupled oscillators.

Data-driven gain computation in the feedback particle filter

A novel method inspired by high-dimensional data-analysis techniques is outlined, based on the time evolution of the particle cloud, which compute values of the gain function for each particle.

Deep FPF: Gain function approximation in high-dimensional setting

A novel approach to approximate the gain function of the feedback particle filter using a stochastic gradient algorithm that allows one to process, in parallel, only a batch of samples ensuring good scaling properties with the number of particles.

Learning techniques for feedback particle filter design

The feedback particle filter (FPF) is an approach to estimating the posterior distribution of the states in a process-observation model that uses Monte Carlo methods to generate and propagate a set of particles, based on the underlying model.

McKean-Vlasov SDEs in nonlinear filtering

A unifying framework is proposed that allows to systematically derive the McKean-Vlasov representations of these filters for the discrete time and continuous time observation case, taking inspiration from the smooth approximation of the data considered in Crisan & Xiong (2010) and Clark & Crisan (2005).

On the continuous time limit of the ensemble Kalman filter

The original Ensemble Kalman Filter algorithm proposed by [1] as well as a recent variant [2] to the respective discretizations are applied and it is shown that in the limit of decreasing stepsize the filter equations converge to an ensemble of interacting (stochastic) differential equations in the ensemble-mean-square sense.

On the Stability and the Uniform Propagation of Chaos of a Class of Extended Ensemble Kalman-Bucy Filters

These stability properties are used to derive uniform and non asymptotic estimates of the propagation of chaos properties of Extended Ensemble Kalman filters, including exponential concentration inequalities.

Analysis of the Ensemble Kalman Filter for Inverse Problems

The goal of this paper is to analyze the EnKF when applied to inverse problems with fixed ensemble size, and to demonstrate that the conclusions of the analysis extend beyond the linear inverse problem setting.