3 papers
cs.LG2021
Generalized Adversarial Distances to Efficiently Discover Classifier Errors
Walter Bennette, Sally Dufek, Karsten Maurer +2
Given a black-box classification model and an unlabeled evaluation dataset from some application domain, efficient strategies need to be developed to evaluate the model. Random sam…
cs.LG2020
Harnessing Adversarial Distances to Discover High-Confidence Errors
Walter Bennette, Karsten Maurer, Sean Sisti
Given a deep neural network image classification model that we treat as a black box, and an unlabeled evaluation dataset, we develop an efficient strategy by which the classifier c…
stat.ML2018
Facility Locations Utility for Uncovering Classifier Overconfidence
Karsten Maurer, Walter Bennette
Assessing the predictive accuracy of black box classifiers is challenging in the absence of labeled test datasets. In these scenarios we may need to rely on a human oracle to evalu…