Video Sources

  • Server 1
Can overfitted deep neural networks in adversarial training generalize? – An approximation viewpoint

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

AnalysisMar. 01, 202452 Min.
Your rating: 0
6 1 vote

What's happening?

Synopsis

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

Similar titles

Periodical
Umhlonyana Ka Mimi
Les 100 vies de Francis Blanche
Making 1899
Seeds, the legacy of the land
Klangwütig – Ein Jahr an der Karajan-Akademie
Journey to Jagdula
In the Hearts of the Living
To Esteban
8. dzień chamsinu
Those Who Wander
Sarah Squared

Leave a comment

Name *
Add a display name
Email *
Your email address will not be published

© 2025 by SOBATLK21 - Nonton Film Streaming Gratis Indonesia | Layarkaca21 INDOXXI Nonton Movie dan TV Series Subtitle Indonesia. All Rights Reserved. Powered by SOBATLK21
x