AdAUC: End-to-end Adversarial AUC Optimization Against Long-tail Problems. (arXiv:2206.12169v1 [cs.LG])

It is well-known that deep learning models are vulnerable to adversarial
examples. Existing studies of adversarial training have made great progress
against this challenge. As a typical trait, they often assume that the class
distribution is overall balanced. However, long-tail datasets are ubiquitous in
a wide spectrum of applications, where the amount of head class instances is
larger than the tail classes. Under such a scenario, AUC is a much more
reasonable metric than accuracy since it is insensitive toward …

arxiv auc lg optimization