Sharp failure rates for the bootstrap particle filter in high dimensions

Abstract

We prove that the maximum of the sample importance weights in a high-dimensional Gaussian particle filter converges to unity unless the ensemble size grows exponentially in the system dimension. Our work is motivated by and parallels the derivations of Bengtsson, Bickel and Li (2007); however, we weaken their assumptions on the eigenvalues of the covariance matrix of the prior distribution and establish rigorously their strong conjecture on when weight collapse occurs. Specifically, we remove the assumption that the nonzero eigenvalues are bounded away from zero, which, although the dimension of the involved vectors grow to infinity, essentially permits the effective system dimension to be bounded. Moreover, with some restrictions on the rate of growth of the maximum eigenvalue, we relax their assumption that the eigenvalues are bounded from above, allowing the system to be dominated by a single mode.

Published in at http://dx.doi.org/10.1214/074921708000000228 the IMS Collections (http://www.imstat.org/publications/imscollections.htm) by the Institute of Mathematical Statistics (http://www.imstat.org)

Cited by in corpus (44)

• Fighting Sample Degeneracy and Impoverishment in Particle Filters: A Review of Intelligent Approaches
• Can local particle filters beat the curse of dimensionality?
• Data Assimilation: A Mathematical Introduction
• A random map implementation of implicit filters
• Multilevel Sequential Monte Carlo Samplers
• A Bayesian approach for inferring neuronal connectivity from calcium fluorescent imaging data
• Particle Filtering with Invertible Particle Flow
• Langevin and Hamiltonian based Sequential MCMC for Efficient Bayesian Filtering in High-dimensional Spaces
• Deterministic Mean-field Ensemble Kalman Filtering
• Implicit particle methods and their connection with variational data assimilation
• Particle filters
• Bayesian inversion in resin transfer molding
• Implicit particle filtering for models with partial noise, and an application to geomagnetic data assimilation
• Sequential Monte Carlo Methods for System Identification
• Conditions for successful data assimilation
• Nested Sequential Monte Carlo Methods
• A Stable Particle Filter in High-Dimensions
• Localization in Ensemble Kalman inversion
• Parameter estimation by implicit sampling
• Stochastic Particle Flow for Nonlinear High-Dimensional Filtering Problems
• On the Stability of Sequential Monte Carlo Methods in High Dimensions
• Ensemble-marginalized Kalman filter for linear time-dependent PDEs with noisy boundary conditions: Application to heat transfer in building walls
• Sequential Monte Carlo with transformations
• Interpolation and Iteration for Nonlinear Filters
• Streaming Variational Monte Carlo
• Error Bounds and Normalizing Constants for Sequential Monte Carlo in High Dimensions
• Multilevel Sequential Monte Carlo Samplers for Normalizing Constants
• Static Parameter Estimation for ABC Approximations of Hidden Markov Models
• Deep FPF: Gain function approximation in high-dimensional setting
• Implicit particle filters for data assimilation
• Kalman filtering and smoothing for linear wave equations with model error
• Bayesian Update with Importance Sampling: Required Sample Size
• Adaptively Blocked Particle Filtering with Spatial Smoothing in Large-Scale Dynamic Random Fields
• Nearly Consistent Finite Particle Estimates in Streaming Importance Sampling
• Quantifying uncertainty in thermal properties of walls by means of Bayesian inversion
• Graph-Based Multiobject Tracking with Embedded Particle Flow
• A High-Dimensional Particle Filter Algorithm
• A New Perspective and Extension of the Gaussian Filter
• Non-Bayesian particle filters
• A Random Walk with Drift: Interview with Peter J. Bickel
• Computational Methods for a Class of Network Models
• Optimal Transportation Methods in Nonlinear Filtering: The feedback particle filter
• Adaptive quantum state tomography with iterative particle filtering
• Reducing the Bias in Blocked Particle Filtering for High-Dimensional Systems