A new no-free-lunch construction and empirical scaling study argue that robust classification can require exponentially more data than standard classification, and that dataset size largely drives certified robust accuracy.
Bartoldson, James Diffenderfer, Konstantinos Parasyris, and Bhavya Kailkhura
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Intriguing Properties of Robust Classification
A new no-free-lunch construction and empirical scaling study argue that robust classification can require exponentially more data than standard classification, and that dataset size largely drives certified robust accuracy.