Salvato in:
Dettagli Bibliografici
Autori principali: Lechner, Tosca, Bie, Alex, Kamath, Gautam
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2509.05137
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918136372002816
author Lechner, Tosca
Bie, Alex
Kamath, Gautam
author_facet Lechner, Tosca
Bie, Alex
Kamath, Gautam
contents We consider the question of learnability of distribution classes in the presence of adaptive adversaries -- that is, adversaries capable of intercepting the samples requested by a learner and applying manipulations with full knowledge of the samples before passing it on to the learner. This stands in contrast to oblivious adversaries, who can only modify the underlying distribution the samples come from but not their i.i.d.\ nature. We formulate a general notion of learnability with respect to adaptive adversaries, taking into account the budget of the adversary. We show that learnability with respect to additive adaptive adversaries is a strictly stronger condition than learnability with respect to additive oblivious adversaries.
format Preprint
id arxiv_https___arxiv_org_abs_2509_05137
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle On the Learnability of Distribution Classes with Adaptive Adversaries
Lechner, Tosca
Bie, Alex
Kamath, Gautam
Machine Learning
We consider the question of learnability of distribution classes in the presence of adaptive adversaries -- that is, adversaries capable of intercepting the samples requested by a learner and applying manipulations with full knowledge of the samples before passing it on to the learner. This stands in contrast to oblivious adversaries, who can only modify the underlying distribution the samples come from but not their i.i.d.\ nature. We formulate a general notion of learnability with respect to adaptive adversaries, taking into account the budget of the adversary. We show that learnability with respect to additive adaptive adversaries is a strictly stronger condition than learnability with respect to additive oblivious adversaries.
title On the Learnability of Distribution Classes with Adaptive Adversaries
topic Machine Learning
url https://arxiv.org/abs/2509.05137