English

Bandit Social Learning with Exploration Episodes

Computer Science and Game Theory 2026-02-06 v1

Abstract

We study a stylized social learning dynamics where self-interested agents collectively follow a simple multi-armed bandit protocol. Each agent controls an ``episode": a short sequence of consecutive decisions. Motivating applications include users repeatedly interacting with an AI, or repeatedly shopping at a marketplace. While agents are incentivized to explore within their respective episodes, we show that the aggregate exploration fails: e.g., its Bayesian regret grows linearly over time. In fact, such failure is a (very) typical case, not just a worst-case scenario. This conclusion persists even if an agent's per-episode utility is some fixed function of the per-round outcomes: e.g., min\min or max\max, not just the sum. Thus, externally driven exploration is needed even when some amount of exploration happens organically.

Keywords

Cite

@article{arxiv.2602.05835,
  title  = {Bandit Social Learning with Exploration Episodes},
  author = {Kiarash Banihashem and Natalie Collina and Aleksandrs Slivkins},
  journal= {arXiv preprint arXiv:2602.05835},
  year   = {2026}
}
R2 v1 2026-07-01T10:22:45.973Z