Sensitivity Analysis for Predictive Uncertainty in Bayesian Neural Networks

TL;DR

This paper introduces a new sensitivity analysis method for Bayesian neural networks that enhances interpretability of predictive uncertainties, distinguishing between epistemic and aleatoric sources, and demonstrates its effectiveness on real-world data.

Contribution

The paper presents a novel sensitivity analysis approach for Bayesian neural networks that improves understanding of uncertainty sources in complex probabilistic models.

Findings

Enhanced interpretability of Bayesian neural networks.

Effective differentiation between epistemic and aleatoric uncertainty.

Validated on real-world datasets.

Abstract

We derive a novel sensitivity analysis of input variables for predictive epistemic and aleatoric uncertainty. We use Bayesian neural networks with latent variables as a model class and illustrate the usefulness of our sensitivity analysis on real-world datasets. Our method increases the interpretability of complex black-box probabilistic models.