We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Existence and Minimax Theorems for Adversarial Surrogate Risks in Binary Classification

Abstract: Adversarial training is one of the most popular methods for training methods robust to adversarial attacks, however, it is not well-understood from a theoretical perspective. We prove and existence, regularity, and minimax theorems for adversarial surrogate risks. Our results explain some empirical observations on adversarial robustness from prior work and suggest new directions in algorithm development. Furthermore, our results extend previously known existence and minimax theorems for the adversarial classification risk to surrogate risks.
Comments: 37 pages, 1 Figure
Subjects: Machine Learning (cs.LG); Statistics Theory (math.ST)
Cite as: arXiv:2206.09098 [cs.LG]
  (or arXiv:2206.09098v1 [cs.LG] for this version)

Submission history

From: Natalie Frank [view email]
[v1] Sat, 18 Jun 2022 03:29:49 GMT (51kb,D)

Link back to: arXiv, form interface, contact.