Can overfitted deep neural networks in adversarial training generalize? – An approximation viewpoint

Опис

In this talk, I will discuss whether overfitted DNNs in adversarial training can generalize from an approximation viewpoint. We prove by construction the existence of infinitely many adversarial training classifiers on over-parameterized DNNs that obtain arbitrarily small adversarial training error (overfitting), whereas achieving good robust generalization error under certain conditions concerning the data quality, well separated, and perturbation level. This construction is optimal and thus points out the fundamental limits of DNNs under adversarial training with statistical guarantees. Part of this talk comes from our recent work.

Деталі

Статус

Випущено

Дата виходу

1 березня 2024 р.

Мова оригіналу

англійська

Країни

Велика Британія

Виробництво

University of Warwick

Бюджет

Немає даних

Збори

Немає даних

Вебсайт

Офіційний сайт