mlsanityData to 13 Oct 2026

Feasibility-Guided Fair Adaptive Offline Reinforcement Learning for Medicaid Care Management

arXiv 2509.09655 ↗First posted 11 Sep 2025cs.LG cs.AI cs.LOcited by 0

On the map: Offline RL and imitation learning from data → Safe offline RL

Abstract (arXiv)

We introduce Feasibility-Guided Fair Adaptive Reinforcement Learning (FG-FARL), an offline RL procedure that calibrates per-group safety thresholds to reduce harm while equalizing a chosen fairness target (coverage or harm) across protected subgroups. Using de-identified longitudinal trajectories from a Medicaid population health management program, we evaluate FG-FARL against behavior cloning (BC) and HACO (Hybrid Adaptive Conformal Offline RL; a global conformal safety baseline). We report off-policy value estimates with bootstrap 95% confidence intervals and subgroup disparity analyses with p-values. FG-FARL achieves comparable value to baselines while improving fairness metrics, demonstrating a practical path to safer and more equitable decision support.

Key facts. No citing papers in the map yet (data through 13 Oct 2026). Placed in Safe offline RL, one of 13 directions in Offline RL and imitation learning from data. Builds on 11 of its 16 references that are in the map. First posted on arXiv 11 Sep 2025.

Builds on

11 of 16 references.

References · 11

  1. 2016Equality of Opportunity in Supervised Learningcited by 2,005
  2. 2020Conservative Q-Learning for Offline Reinforcement Learningcited by 1,782
  3. 2021Offline Reinforcement Learning with Implicit Q-Learningcited by 1,257
  4. 2017Constrained Policy Optimizationcited by 931
  5. 2021A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantificationcited by 786
  6. 2018Reward Constrained Policy Optimizationcited by 400
  7. 2016Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learningcited by 285
  8. 2019Batch Policy Learning under Constraintscited by 243
  9. 2018Safe Exploration in Continuous Action Spacescited by 240
  10. 2019Minimax Weight and Q-Function Learning for Off-Policy Evaluationcited by 150
  11. 2016Fairness in Reinforcement Learningcited by 93

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Hybrid Adaptive Conformal Offline Reinforcement Learning for Fair Population Health Management
  2. 2025Test-Time Learning and Inference-Time Deliberation for Efficiency-First Offline Reinforcement Learning in Care Coordination and Population Health Management
  3. 2025PyCFRL: A Python library for counterfactually fair offline reinforcement learning via sequential data preprocessing
  4. 2026A Distribution Mapping Approach to Counterfactually Fair Reinforcement Learning
  5. 2022Offline Policy Evaluation and Optimization under Confounding
  6. 2024Reinforcement Learning in Dynamic Treatment Regimes Needs Critical Reexamination
  7. 2025Offline Guarded Safe Reinforcement Learning for Medical Treatment Optimization Strategies
  8. 2022Semi-Markov Offline Reinforcement Learning for Healthcare