Sensitivity Analysis for Predictive Uncertainty in Bayesian Neural Networks
This work addresses interpretability for users of complex probabilistic models, though it appears incremental as it builds on existing Bayesian neural network frameworks.
The authors tackled the problem of interpreting predictive uncertainty in Bayesian neural networks by developing a sensitivity analysis method for input variables, which they demonstrated on real-world datasets to increase model interpretability.
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.