Multilevel Bootstrap Particle Filter

TL;DR

This paper introduces a multilevel bootstrap particle filter that significantly reduces computation time in sequential Monte Carlo methods while maintaining theoretical guarantees, with up to 85% efficiency gains demonstrated in experiments.

Contribution

The paper proposes a novel multilevel bootstrap particle filter that retains key theoretical properties and offers substantial computational efficiency improvements in specific applications.

Findings

Up to 85% reduction in computation time.

Retains strong law of large numbers and CLT.

Efficiency gains are application-dependent.

Abstract

We consider situations where the applicability of sequential Monte Carlo particle filters is compromised due to the expensive evaluation of the particle weights. To alleviate this problem, we propose a new particle filter algorithm based on the multilevel approach. We show that the resulting multilevel bootstrap particle filter (MLBPF) retains the strong law of large numbers as well as the central limit theorem of classical particle filters under mild conditions. Our numerical experiments demonstrate up to 85\% reduction in computation time compared to the classical bootstrap particle filter, in certain settings. While it should be acknowledged that this reduction is highly application dependent, and a similar gain should not be expected for all applications across the board, we believe that this substantial improvement in certain settings makes MLBPF an important addition to the family of sequential Monte Carlo methods.