An overview of differentiable particle filters for data-adaptive sequential Bayesian inference

TL;DR

This paper reviews recent advances in differentiable particle filters, highlighting their design choices and applications in complex, high-dimensional sequential data inference tasks, leveraging neural networks and gradient-based optimization.

Contribution

It provides a comprehensive overview of the design options and recent developments in differentiable particle filters for data-adaptive Bayesian inference.

Findings

Differentiable particle filters enhance inference in high-dimensional tasks.

Neural network components improve flexibility and performance.

Various design choices impact effectiveness and efficiency.

Abstract

By approximating posterior distributions with weighted samples, particle filters (PFs) provide an efficient mechanism for solving non-linear sequential state estimation problems. While the effectiveness of particle filters has been recognised in various applications, their performance relies on the knowledge of dynamic models and measurement models, as well as the construction of effective proposal distributions. An emerging trend involves constructing components of particle filters using neural networks and optimising them by gradient descent, and such data-adaptive particle filtering approaches are often called differentiable particle filters. Due to the expressiveness of neural networks, differentiable particle filters are a promising computational tool for performing inference on sequential data in complex, high-dimensional tasks, such as vision-based robot localisation. In this paper, we review recent advances in differentiable particle filters and their applications. We place special emphasis on different design choices for key components of differentiable particle filters, including dynamic models, measurement models, proposal distributions, optimisation objectives, and differentiable resampling techniques.