English

Elephant Random Walk with multiple extractions

Probability 2025-07-10 v1

Abstract

Consider a generalized Elephant Random Walk in which the step is chosen by selecting kk previous steps with kk odd and then going in the majority direction with a probability pp and in the opposite direction otherwise. In the k=1k=1 case the model is the original one and could be resolved exactly by analogy with Friedman's urn. However the analogy cannot be extended to the k>2k>2 case already. In this paper we show how to treat the model for each kk by analogy with the more general urn model of Hill, Lane and Sudderth. Interestingly for k>2k>2 we found a critical dependence from the initial conditions beyond a certain values of the memory parameter pp, and regions of convergence with entropy that is sub-linear in the number of steps.

Keywords

Cite

@article{arxiv.2507.06478,
  title  = {Elephant Random Walk with multiple extractions},
  author = {Simone Franchini},
  journal= {arXiv preprint arXiv:2507.06478},
  year   = {2025}
}

Comments

This paper serves as a handout for the homonymous talk in the session CS19 Reinforcement Models: Elephant Random Walk, 44th Conference on Stochastic Processes and their Applications, Wroclaw, July 14-18, 2025. 17 pages, 4 figures. Replaces arXiv:2210.12585v1

R2 v1 2026-07-01T03:52:33.507Z