Provable Tradeoffs in Adversarially Robust Classification-Reference-Cited by-同舟云学术

Provable Tradeoffs in Adversarially Robust Classification

Author:

Dobriban Edgar¹^ORCID,Hassani Hamed²^ORCID,Hong David¹^ORCID,Robey Alexander²^ORCID

Affiliation:

1. Department of Statistics and Data Science, University of Pennsylvania, Philadelphia, PA, USA

2. Department of Electrical and Systems Engineering, University of Pennsylvania, Philadelphia, PA, USA

Funder

NSF BIGDATA Grant

NSF-Simons Award on the Mathematical and Scientific Foundations of Deep Learning

Dean’s Fund for Post-Doctoral Research of the Wharton School

NSF Mathematical Sciences Post-Doctoral Research Fellowship

NSF HDR TRIPODS

NSF

NSF CAREER Award

Air Force Office of Scientific Research Young Investigator Program

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Subject

Library and Information Sciences,Computer Science Applications,Information Systems

Link

Reference74 articles.

1. Understanding deep learning requires rethinking generalization;zhang;Proc Int Conf Learn Represent,2017

2. Robustness for non-parametric classification: A generic attack and defense;yang;Proc 23rd Int Conf Artif Intell Statist,2020

4. When are non-parametric methods robust?;bhattacharjee;Proc 37th Int Conf Mach Learn,2020

5. Adversarial training can hurt generalization;raghunathan;Proc Workshop Identifying Understand Deep Learn Phenomena (ICML),2019