We introduce a novel framework of ranking with abstention, where the lea...
In many applications, the labeled data at the learner's disposal is subj...
Cross-entropy is a widely used loss function in applications. It coincid...
We present a detailed study of estimation errors in terms of surrogate l...
We present a more general analysis of H-calibration for adversarially
ro...
Adversarial robustness is an increasingly critical property of classifie...
A novel solve-training framework is proposed to train neural network in
...