We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computer Vision and Pattern Recognition

Title: On the Practicality of Deterministic Epistemic Uncertainty

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.
Comments: International Conference on Machine Learning 2022
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Cite as: arXiv:2107.00649 [cs.CV]
  (or arXiv:2107.00649v3 [cs.CV] for this version)

Submission history

From: Janis Postels [view email]
[v1] Thu, 1 Jul 2021 17:59:07 GMT (31528kb,D)
[v2] Tue, 13 Jul 2021 19:49:32 GMT (31529kb,D)
[v3] Tue, 5 Jul 2022 07:39:17 GMT (31678kb,D)

Link back to: arXiv, form interface, contact.