Revisiting semi-supervised training objectives for differentiable particle filters

TL;DR

This paper evaluates semi-supervised training methods for differentiable particle filters, demonstrating their effectiveness in scenarios with limited labeled data through experiments in simulated environments.

Contribution

It compares two semi-supervised training objectives for differentiable particle filters, highlighting their performance when labeled data is scarce.

Findings

Semi-supervised methods improve performance with limited labels

Differentiable particle filters can be effectively trained with semi-supervised objectives

Results show promising application in simulated environments

Abstract

Differentiable particle filters combine the flexibility of neural networks with the probabilistic nature of sequential Monte Carlo methods. However, traditional approaches rely on the availability of labelled data, i.e., the ground truth latent state information, which is often difficult to obtain in real-world applications. This paper compares the effectiveness of two semi-supervised training objectives for differentiable particle filters. We present results in two simulated environments where labelled data are scarce.