On biased random walks, corrupted intervals, and learning under adversarial design

Daniel Berend, Aryeh Kontorovich, Lev Reyzin, Thomas Robinson

Research output: Contribution to journalArticlepeer-review

Abstract

We tackle some fundamental problems in probability theory on corrupted random processes on the integer line. We analyze when a biased random walk is expected to reach its bottommost point and when intervals of integer points can be detected under a natural model of noise. We apply these results to problems in learning thresholds and intervals under a new model for learning under adversarial design.

Original languageEnglish
Pages (from-to)887-905
Number of pages19
JournalAnnals of Mathematics and Artificial Intelligence
Volume88
Issue number8
DOIs
StatePublished - 1 Aug 2020

Keywords

  • Adversarial learning
  • Classification noise
  • Random walks

ASJC Scopus subject areas

  • Artificial Intelligence
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'On biased random walks, corrupted intervals, and learning under adversarial design'. Together they form a unique fingerprint.

Cite this