Uncertainty Disentanglement with Non-stationary Heteroscedastic Gaussian Processes for Active Learning

TL;DR

This paper introduces a non-stationary heteroscedastic Gaussian process model that disentangles aleatoric and epistemic uncertainties, enhancing interpretability and active learning performance.

Contribution

It proposes a novel Gaussian process model that separates different uncertainty types and can be trained with gradient-based methods, improving interpretability and active learning applications.

Findings

Effective separation of uncertainties demonstrated on multiple datasets

Improved active learning performance using epistemic uncertainty

Model trained efficiently with gradient-based optimization

Abstract

Gaussian processes are Bayesian non-parametric models used in many areas. In this work, we propose a Non-stationary Heteroscedastic Gaussian process model which can be learned with gradient-based techniques. We demonstrate the interpretability of the proposed model by separating the overall uncertainty into aleatoric (irreducible) and epistemic (model) uncertainty. We illustrate the usability of derived epistemic uncertainty on active learning problems. We demonstrate the efficacy of our model with various ablations on multiple datasets.