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

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 p...

Full description

Saved in:
Bibliographic Details
Published inAnnals of mathematics and artificial intelligence Vol. 88; no. 8; pp. 887 - 905
Main Authors Berend, Daniel, Kontorovich, Aryeh, Reyzin, Lev, Robinson, Thomas
Format Journal Article
LanguageEnglish
Published Cham Springer International Publishing 01.08.2020
Springer
Springer Nature B.V
Subjects
Artificial Intelligence
Complex Systems
Computer Science
Distance learning
Hypotheses
Integers
Intervals
Learning
Mathematics
Probability
Probability theory
Random processes
Random variables
Random walk
Regular Submission
Classification noise
Primary 60G50
68T05
Adversarial learning
Random walks
Secondary 68Q32
Online AccessGet full text

Cover

More Information
Summary: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.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1012-2443
1573-7470
DOI:10.1007/s10472-020-09696-1