mlsanityData to 14 Oct 2026

HR-Bandit: Human-AI Collaborated Linear Recourse Bandit

arXiv 2410.14640 ↗First posted 18 Oct 2024cs.LGcited by 2

Not on the map. Why?

Abstract (arXiv)

Human doctors frequently recommend actionable recourses that allow patients to modify their conditions to access more effective treatments. Inspired by such healthcare scenarios, we propose the Recourse Linear UCB (textsfRLinUCB) algorithm, which optimizes both action selection and feature modifications by balancing exploration and exploitation. We further extend this to the Human-AI Linear Recourse Bandit (textsfHR-Bandit), which integrates human expertise to enhance performance. textsfHR-Bandit offers three key guarantees: (i) a warm-start guarantee for improved initial performance, (ii) a human-effort guarantee to minimize required human interactions, and (iii) a robustness guarantee that ensures sublinear regret even when human decisions are suboptimal. Empirical results, including a healthcare case study, validate its superior performance against existing benchmarks.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 36 of its 65 references that are in the map. First posted on arXiv 18 Oct 2024.

Builds on

36 of 65 references.

References · 36

  1. 2017Explanation in Artificial Intelligence: Insights from the Social Sciencescited by 1,206
  2. 2017Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPRcited by 1,089
  3. 2019Introduction to Multi-Armed Banditscited by 481
  4. 2018Actionable Recourse in Linear Classificationcited by 328
  5. 2013Learning to Optimize Via Posterior Samplingcited by 256
  6. 2017Predict Responsibly: Improving Fairness and Accuracy by Learning to Defercited by 233
  7. 2019Interpretable Counterfactual Explanations Guided by Prototypescited by 223
  8. 2019Model-Agnostic Counterfactual Explanations for Consequential Decisionscited by 204
  9. 2020Multi-Objective Counterfactual Explanationscited by 157
  10. 2019Learning Model-Agnostic Counterfactual Explanations for Tabular Datacited by 146
  11. 2019Preserving Causal Constraints in Counterfactual Explanations for Machine Learning Classifierscited by 144
  12. 2020Algorithmic recourse under imperfect causal knowledge: a probabilistic approachcited by 140
24 more
  1. 2019Towards Realistic Individual Recourse and Actionable Explanations in Black-Box Decision Making Systemscited by 131
  2. 2020Learning to Complement Humanscited by 124
  3. 2020A survey of algorithmic recourse: definitions, formulations, solutions, and prospectscited by 118
  4. 2021Towards Robust and Reliable Algorithmic Recoursecited by 102
  5. 2019The Algorithmic Automation Problem: Prediction, Triage, and Human Effortcited by 101
  6. 2015Global Convergence of ADMM in Nonconvex Nonsmooth Optimizationcited by 98
  7. 2020On the Fairness of Causal Algorithmic Recoursecited by 70
  8. 2014Convergence Analysis of Alternating Direction Method of Multipliers for a Family of Nonconvex Problemscited by 64
  9. 2021On the Adversarial Robustness of Causal Algorithmic Recoursecited by 56
  10. 2020Algorithmic Recourse in the Wild: Understanding the Impact of Data and Model Shiftscited by 36
  11. 2021Human-AI Collaboration with Bandit Feedbackcited by 36
  12. 2023Looping in the Human Collaborative and Explainable Bayesian Optimizationcited by 27
  13. 2018Learning under selective labels in the presence of expert consistencycited by 18
  14. 2022The Best Decisions Are Not the Best Advice: Making Adherence-Aware Recommendationscited by 16
  15. 2021Improving Human Sequential Decision-Making with Reinforcement Learningcited by 12
  16. 2020A Bandit Model for Human-Machine Decision Making with Private Information and Opacitycited by 10
  17. 2022Hierarchical Conversational Preference Elicitation with Bandit Feedbackcited by 10
  18. 2019A Case for Backward Compatibility for Human-AI Teamscited by 7
  19. 2022Leveraging Initial Hints for Free in Stochastic Linear Banditscited by 7
  20. 2023Learning Complementary Policies for Human-AI Teamscited by 7
  21. 2023BO-Muse: A human expert and AI teaming framework for accelerated experimental designcited by 6
  22. 2024Principled Bayesian Optimisation in Collaboration with Human Expertscited by 5
  23. 2023Confounding-Robust Policy Improvement with Human-AI Teamscited by 4
  24. 2023Contextual Bandits with Budgeted Information Revealcited by 3

Built on by

Cited by year

Citing papers per year: 2025: 2.

2 papers.

  1. 2025Pre-Trained AI Model Assisted Online Decision-Making under Missing Covariates: A Theoretical Perspectivecited by 3
  2. 2025Learning with Incomplete Context: Linear Contextual Bandits with Pretrained Imputationcited by 0