Poster

On the Practicality of Deterministic Epistemic Uncertainty

Janis Postels · Mattia Seg├╣ · Tao Sun · Luca Daniel Sieber · Luc Van Gool · Fisher Yu · Federico Tombari

Hall E #506

Keywords: [ PM: Bayesian Models and Methods ] [ MISC: Supervised Learning ] [ APP: Computer Vision ] [ MISC: Representation Learning ] [ DL: Robustness ]

Abstract:

A set of novel approaches for estimating epistemic uncertainty in deep neural networks with a single forward pass has recently emerged as a valid alternative to Bayesian Neural Networks. On the premise of informative representations, these deterministic uncertainty methods (DUMs) achieve strong performance on detecting out-of-distribution (OOD) data while adding negligible computational costs at inference time. However, it remains unclear whether DUMs are well calibrated and can seamlessly scale to real-world applications - both prerequisites for their practical deployment. To this end, we first provide a taxonomy of DUMs, and evaluate their calibration under continuous distributional shifts. Then, we extend them to semantic segmentation. We find that, while DUMs scale to realistic vision tasks and perform well on OOD detection, the practicality of current methods is undermined by poor calibration under distributional shifts.

Chat is not available.