Bayesian Inference in Recurrent Explicit Duration Switching Linear Dynamical Systems

TL;DR

This paper introduces REDSLDS, a new model that enhances segmentation in dynamical systems by integrating recurrent explicit duration variables and Pólya-gamma augmentation for improved inference.

Contribution

The paper presents a novel REDSLDS model with recurrent explicit duration variables and a Pólya-gamma based inference scheme, advancing segmentation performance in dynamical systems.

Findings

Improved segmentation on benchmark datasets

Effective incorporation of explicit duration variables

Enhanced inference scheme with Pólya-gamma augmentation

Abstract

In this paper, we propose a novel model called Recurrent Explicit Duration Switching Linear Dynamical Systems (REDSLDS) that incorporates recurrent explicit duration variables into the rSLDS model. We also propose an inference and learning scheme that involves the use of P\'olya-gamma augmentation. We demonstrate the improved segmentation capabilities of our model on three benchmark datasets, including two quantitative datasets and one qualitative dataset.