Search Results for author:
Found 2 papers, 1 papers with code
Beyond Distribution Shift: Spurious Features Through the Lens of Training Dynamics
(3) We empirically show that the harmful spurious features can be detected by observing the learning dynamics of the DNN's early layers.
Augmentation by Counterfactual Explanation -- Fixing an Overconfident Classifier
The classification outcome should reflect a high uncertainty on ambiguous in-distribution samples that lie close to the decision boundary.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
