mlsanityData to 7 Oct 2026

A Bit Better? Quantifying Information for Bandit Learning

arXiv 2102.09488 ↗First posted 18 Feb 2021cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

The information ratio offers an approach to assessing the efficacy with which an agent balances between exploration and exploitation. Originally, this was defined to be the ratio between squared expected regret and the mutual information between the environment and action-observation pair, which represents a measure of information gain. Recent work has inspired consideration of alternative information measures, particularly for use in analysis of bandit learning algorithms to arrive at tighter regret bounds. We investigate whether quantification of information via such alternatives can improve the realized performance of information-directed sampling, which aims to minimize the information ratio.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 16 of its 22 references that are in the map. First posted on arXiv 18 Feb 2021.

Builds on

16 of 22 references.

References · 16

  1. 2017A Tutorial on Thompson Samplingcited by 420
  2. 2013Learning to Optimize Via Posterior Samplingcited by 253
  3. 2014An Information-Theoretic Analysis of Thompson Samplingcited by 178
  4. 2014Learning to Optimize via Information-Directed Samplingcited by 168
  5. 2018Information Directed Sampling and Bandits with Heteroscedastic Noisecited by 75
  6. 2019Information-Theoretic Confidence Bounds for Reinforcement Learningcited by 51
  7. 2019An Information-Theoretic Approach to Minimax Regret in Partial Monitoringcited by 41
  8. 2020Information Directed Sampling for Linear Partial Monitoringcited by 39
  9. 2020Mirror Descent and the Information Ratiocited by 39
  10. 2018Satisficing in Time-Sensitive Bandit Learningcited by 38
  11. 2018An Information-Theoretic Analysis for Thompson Sampling with Many Actionscited by 35
  12. 2020Improved Regret for Zeroth-Order Adversarial Bandit Convex Optimisationcited by 33
4 more
  1. 2013Prior-free and prior-dependent regret bounds for Thompson Samplingcited by 25
  2. 2019Connections Between Mirror Descent, Thompson Sampling and the Information Ratiocited by 23
  3. 2019Exploration by Optimisation in Partial Monitoringcited by 20
  4. 2015Bandit Convex Optimization: sqrt{T} Regret in One Dimensioncited by 10

Built on by

Cited by year

Citing papers per year: 2021: 2; 2022: 2; 2023: 1.

5 papers.

  1. 2021Reinforcement Learning, Bit by Bitcited by 59
  2. 2022Minimax Regret for Partial Monitoring: Infinite Outcomes and Rustichini's Regretcited by 13
  3. 2021Bandit Phase Retrievalcited by 12
  4. 2022Gaussian Imagination in Bandit Learningcited by 5
  5. 2023Efficient Online Decision Tree Learning with Active Feature Acquisitioncited by 2