Active Learning in Gaussian Process State Space Model

TL;DR

This paper explores active learning strategies for Gaussian Process state-space models, aiming to efficiently learn system dynamics by selecting the most informative inputs through mutual information criteria.

Contribution

It introduces two novel methods for approximating mutual information in GPSSMs to improve active learning of non-linear system dynamics.

Findings

Effective active learning approaches demonstrated in physical systems.

Improved system dynamics modeling with fewer data points.

Validation of mutual information approximation methods.

Abstract

We investigate active learning in Gaussian Process state-space models (GPSSM). Our problem is to actively steer the system through latent states by determining its inputs such that the underlying dynamics can be optimally learned by a GPSSM. In order that the most informative inputs are selected, we employ mutual information as our active learning criterion. In particular, we present two approaches for the approximation of mutual information for the GPSSM given latent states. The proposed approaches are evaluated in several physical systems where we actively learn the underlying non-linear dynamics represented by the state-space model.