Sampling-Free Probabilistic Deep State-Space Models

TL;DR

This paper introduces a deterministic inference algorithm for Probabilistic Deep State-Space Models, enabling efficient training and testing of complex dynamical systems with neural network-based transition and emission models.

Contribution

It presents the first deterministic inference method for ProDSSMs, improving computational efficiency and performance in modeling unknown dynamical systems.

Findings

Efficient training and testing of ProDSSMs achieved.

Superior predictive performance over existing methods.

Reduced computational cost while maintaining accuracy.

Abstract

Many real-world dynamical systems can be described as State-Space Models (SSMs). In this formulation, each observation is emitted by a latent state, which follows first-order Markovian dynamics. A Probabilistic Deep SSM (ProDSSM) generalizes this framework to dynamical systems of unknown parametric form, where the transition and emission models are described by neural networks with uncertain weights. In this work, we propose the first deterministic inference algorithm for models of this type. Our framework allows efficient approximations for training and testing. We demonstrate in our experiments that our new method can be employed for a variety of tasks and enjoys a superior balance between predictive performance and computational budget.