The Peril of Popular Deep Learning Uncertainty Estimation Methods

Understand

Uncertainty estimation (UE) techniques -- such as the Gaussian process (GP), Bayesian neural networks (BNN), Monte Carlo dropout (MCDropout) -- aim to improve the interpretability of machine learning models by assigning an estimated uncertainty value to each of their prediction outputs.

• However, since too high uncertainty estimates can have fatal consequences in practice, this paper analyzes the above techniques.
• Firstly, we show that GP methods always yield high uncertainty estimates on out of distribution (OOD) data.
• Secondly, we show on a 2D toy example that both BNNs and MCDropout do not give high uncertainty estimates on OOD samples.