PyDPF: A Python Package for Differentiable Particle Filtering

TL;DR

PyDPF introduces a unified Python framework for differentiable particle filtering, enabling easier implementation, comparison, and application of advanced state-space modeling techniques in time series analysis.

Contribution

This work provides the first unified API for differentiable particle filters in PyTorch, facilitating broader research and application in state-space modeling.

Findings

Validated by reproducing existing experiments

Demonstrated application to common state space challenges

Facilitated comparison of differentiable particle filtering methods

Abstract

State-space models (SSMs) are a widely used tool in time series analysis. In the complex systems that arise from real-world data, it is common to employ particle filtering (PF), an efficient Monte Carlo method for estimating the hidden state corresponding to a sequence of observations. Applying particle filtering requires specifying both the parametric form and the parameters of the system, which are often unknown and must be estimated. Gradient-based optimisation techniques cannot be applied directly to standard particle filters, as the filters themselves are not differentiable. However, several recently proposed methods modify the resampling step to make particle filtering differentiable. In this paper, we present an implementation of several such differentiable particle filters (DPFs) with a unified API built on the popular PyTorch framework. Our implementation makes these algorithms easily accessible to a broader research community and facilitates straightforward comparison between them. We validate our framework by reproducing experiments from several existing studies and demonstrate how DPFs can be applied to address several common challenges with state space modelling.