AI Chat AI Image Generator AI Video Text to Speech

VC Classes are Adversarially Robustly Learnable, but Only Improperly

02/12/2019

∙

by Omar Montasser, et al.

∙

∙

We study the question of learning an adversarially robust predictor. We show that any hypothesis class H with finite VC dimension is robustly PAC learnable with an improper learning rule. The requirement of being improper is necessary as we exhibit examples of hypothesis classes H with finite VC dimension that are not robustly PAC learnable with any proper learning rule.

Omar Montasser
12 publications
Steve Hanneke
42 publications
Nathan Srebro
96 publications

page 1

page 2

page 3

page 4

research

∙ 11/10/2022

Probabilistically Robust PAC Learning

Recently, Robey et al. propose a notion of probabilistic robustness, whi...

0 Vinod Raman, et al. ∙

research

∙ 03/01/2021

Robust learning under clean-label attack

We study the problem of robust learning under clean-label data-poisoning...

13 Avrim Blum, et al. ∙

research

∙ 06/26/2022

Adversarially Robust Learning of Real-Valued Functions

We study robustness to test-time adversarial attacks in the regression s...

0 Idan Attias, et al. ∙

research

∙ 08/21/2023

Fat Shattering, Joint Measurability, and PAC Learnability of POVM Hypothesis Classes

We characterize learnability for quantum measurement classes by establis...

0 Abram Magner, et al. ∙

research

∙ 09/19/2013

Predictive PAC Learning and Process Decompositions

We informally call a stochastic process learnable if it admits a general...

0 Cosma Rohilla Shalizi, et al. ∙

research

∙ 03/15/2022

Approximability and Generalisation

Approximate learning machines have become popular in the era of small de...

0 Andrew J. Turner, et al. ∙

research

∙ 08/11/2023

On the equivalence of Occam algorithms

Blumer et al. (1987, 1989) showed that any concept class that is learnab...

0 Zaman Keinath-Esmail, et al. ∙