English

Best-Arm Identification with Noisy Actuation

Information Theory 2026-04-03 v1 Machine Learning math.IT

Abstract

In this paper, we consider a multi-armed bandit (MAB) instance and study how to identify the best arm when arm commands are conveyed from a central learner to a distributed agent over a discrete memoryless channel (DMC). Depending on the agent capabilities, we provide communication schemes along with their analysis, which interestingly relate to the zero-error capacity of the underlying DMC.

Keywords

Cite

@article{arxiv.2604.02255,
  title  = {Best-Arm Identification with Noisy Actuation},
  author = {Merve Karakas and Osama Hanna and Lin F. Yang and Christina Fragouli},
  journal= {arXiv preprint arXiv:2604.02255},
  year   = {2026}
}
R2 v1 2026-07-01T11:51:29.588Z