Trust Me, I Know the Way: Predictive Uncertainty in the Presence of Shortcut Learning
Track: tiny / short paper (up to 3 pages)
Keywords: uncertainty quantification, entropy, shortcut learning
TL;DR: Shortcut learning is decisive for epistemic uncertainty manifesting as disagreement between conflicting hypotheses.
Abstract: The correct way to quantify predictive uncertainty in neural networks remains a topic of active discussion.
In particular, it is unclear whether the state-of-the art entropy decomposition leads to a meaningful representation of model, or *epistemic*, uncertainty (EU) in the light of a debate that pits *ignorance* against *disagreement* perspectives.
We aim to reconcile the conflicting viewpoints by arguing that both are valid but arise from different learning situations.
Notably, we show that the presence of *shortcuts* is decisive for EU manifesting as disagreement.
Anonymization: This submission has been anonymized for double-blind review via the removal of identifying information such as names, affiliations, and identifying URLs.
Format: Yes, the presenting author will attend in person if this work is accepted to the workshop.
Funding: No, the presenting author of this submission does *not* fall under ICLR’s funding aims, or has sufficient alternate funding.
Presenter: ~Lisa_Wimmer1
Submission Number: 19
Loading