Saved in:
Bibliographic Details
Main Authors: Hirling, Annika, Nicoletti, Giorgio, Celani, Antonio
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.20096
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915692365742080
author Hirling, Annika
Nicoletti, Giorgio
Celani, Antonio
author_facet Hirling, Annika
Nicoletti, Giorgio
Celani, Antonio
contents The Multi-Armed Bandit problem provides a fundamental framework for analyzing the tension between exploration and exploitation in sequential learning. This paper explores Information Directed Sampling (IDS) policies, a class of heuristics that balance immediate regret against information gain. We focus on the tractable environment of two-state Bernoulli bandits as a minimal model to rigorously compare heuristic strategies against the optimal policy. We extend the IDS framework to the discounted infinite-horizon setting by introducing a modified information measure and a tuning parameter to modulate the decision-making behavior. We examine two specific problem classes: symmetric bandits and the scenario involving one fair coin. In the symmetric case we show that IDS achieves bounded cumulative regret, whereas in the one-fair-coin scenario the IDS policy yields a regret that scales logarithmically with the horizon, in agreement with classical asymptotic lower bounds. This work serves as a pedagogical synthesis, aiming to bridge concepts from reinforcement learning and information theory for an audience of statistical physicists.
format Preprint
id arxiv_https___arxiv_org_abs_2512_20096
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Information-directed sampling for bandits: a primer
Hirling, Annika
Nicoletti, Giorgio
Celani, Antonio
Machine Learning
Information Theory
The Multi-Armed Bandit problem provides a fundamental framework for analyzing the tension between exploration and exploitation in sequential learning. This paper explores Information Directed Sampling (IDS) policies, a class of heuristics that balance immediate regret against information gain. We focus on the tractable environment of two-state Bernoulli bandits as a minimal model to rigorously compare heuristic strategies against the optimal policy. We extend the IDS framework to the discounted infinite-horizon setting by introducing a modified information measure and a tuning parameter to modulate the decision-making behavior. We examine two specific problem classes: symmetric bandits and the scenario involving one fair coin. In the symmetric case we show that IDS achieves bounded cumulative regret, whereas in the one-fair-coin scenario the IDS policy yields a regret that scales logarithmically with the horizon, in agreement with classical asymptotic lower bounds. This work serves as a pedagogical synthesis, aiming to bridge concepts from reinforcement learning and information theory for an audience of statistical physicists.
title Information-directed sampling for bandits: a primer
topic Machine Learning
Information Theory
url https://arxiv.org/abs/2512.20096