mlsanityData to 7 Oct 2026

Conformal Off-Policy Prediction in Contextual Bandits

arXiv 2206.04405 ↗First posted 9 Jun 2022stat.ML cs.LGcited by 25

Not on the map. Why?

Abstract (arXiv)

Most off-policy evaluation methods for contextual bandits have focused on the expected outcome of a policy, which is estimated via methods that at best provide only asymptotic guarantees. However, in many applications, the expectation may not be the best measure of performance as it does not capture the variability of the outcome. In addition, particularly in safety-critical settings, stronger guarantees than asymptotic correctness may be required. To address these limitations, we consider a novel application of conformal prediction to contextual bandits. Given data collected under a behavioral policy, we propose conformal off-policy prediction (COPP), which can output reliable predictive intervals for the outcome under a new target policy. We provide theoretical finite-sample guarantees without making any additional assumptions beyond the standard contextual bandit setup, and empirically demonstrate the utility of COPP compared with existing methods on synthetic and real-world data.

Key facts. Cited by 25 papers in the map (data through 7 Oct 2026). Builds on 22 of its 39 references that are in the map. First posted on arXiv 9 Jun 2022.

Builds on

22 of 39 references.

References · 22

  1. 2021A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantificationcited by 777
  2. 2017A Distributional Perspective on Reinforcement Learningcited by 743
  3. 2019Conformalized Quantile Regressioncited by 620
  4. 2019Conformal Prediction Under Covariate Shiftcited by 511
  5. 2015Doubly Robust Policy Evaluation and Optimizationcited by 158
  6. 2020Conformal Inference of Counterfactuals and Individual Treatment Effectscited by 150
  7. 2018More Robust Doubly Robust Off-policy Evaluationcited by 141
  8. 2013Introducing LETOR 4.0 Datasetscited by 120
  9. 2019Towards Optimal Off-Policy Evaluation for Reinforcement Learning with Marginalized Importance Samplingcited by 118
  10. 2016Optimal and Adaptive Off-policy Evaluation in Contextual Banditscited by 111
  11. 2021Learning Optimal Conformal Classifierscited by 108
  12. 2016Off-policy evaluation for slate recommendationcited by 102
10 more
  1. 2019Doubly robust off-policy evaluation with shrinkagecited by 87
  2. 2018A Contextual Bandit Bake-offcited by 64
  3. 2019Being Optimistic to Be Conservative: Quickly Learning a CVaR Policycited by 63
  4. 2021Sensitivity Analysis of Individual Treatment Effects: A Robust Conformal Inference Approachcited by 62
  5. 2021Universal Off-Policy Evaluationcited by 43
  6. 2021Conformal Sensitivity Analysis for Individual Treatment Effectscited by 40
  7. 2020Confident Off-Policy Evaluation and Selection through Self-Normalized Importance Weightingcited by 38
  8. 2021Off-Policy Risk Assessment in Contextual Banditscited by 29
  9. 2018Best Arm Identification for Contaminated Banditscited by 23
  10. 2020Learning Robust Decision Policies from Observational Datacited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 4; 2023: 5; 2024: 7; 2025: 2; 2026: 7.

25 papers.

Sort:

By area:

  1. 2022Probabilistic Conformal Prediction Using Conditional Random Samplescited by 47
  2. 2023Conformal Decision Theory: Safe Autonomous Decisions from Imperfect Predictionscited by 46
  3. 2023Conformal Meta-learners for Predictive Inference of Individual Treatment Effectscited by 23
  4. 2024Conformal Prediction for Causal Effects of Continuous Treatmentscited by 19
  5. 2022Conformal Off-policy Predictioncited by 14
  6. 2024Conformalized Interactive Imitation Learning: Handling Expert Shift and Intermittent Feedbackcited by 11
  7. 2024Conformal Counterfactual Inference under Hidden Confoundingcited by 10
  8. 2023Conformal Off-Policy Evaluation in Markov Decision Processescited by 6
  9. 2024What If We Had Used a Different App? Reliable Counterfactual KPI Analysis in Wireless Systemscited by 5
  10. 2026Conformal Policy Controlcited by 5
  11. 2022Distribution-Free Finite-Sample Guarantees and Split Conformal Predictioncited by 4
  12. 2024Conformalized Teleoperation: Confidently Mapping Human Inputs to High-Dimensional Robot Actionscited by 4
13 more
  1. 2023When is Off-Policy Evaluation (Reward Modeling) Useful in Contextual Bandits? A Data-Centric Perspectivecited by 3
  2. 2024Conformal Off-Policy Prediction for Multi-Agent Systemscited by 3
  3. 2024Efficient Online Set-valued Classification with Bandit Feedbackcited by 3
  4. 2025PERRY: Policy Evaluation with Confidence Intervals using Auxiliary Datacited by 3
  5. 2023Off-Policy Evaluation with Out-of-Sample Guaranteescited by 2
  6. 2025Doubly Robust Fusion of Many Treatments for Policy Learningcited by 1
  7. 2022Bayesian Counterfactual Mean Embeddings and Off-Policy Evaluationcited by 0
  8. 2026AdaptNC: Adaptive Nonconformity Scores for Conformal Prediction under Distribution Shiftcited by 0
  9. 2026Set-Valued Policy Learningcited by 0
  10. 2026Safety Screening for Voltage Control in Active Distribution Grids via Distributionally Robust Conformal Screeningcited by 0
  11. 2026Certified Safety Curation: Distribution-Free Guarantees for Safe Offline Reinforcement Learningcited by 0
  12. 2026Model-based Bootstrap for Offline Policy Evaluation in Tabular Reinforcement Learningcited by 0
  13. 2026Counterfactual Online Conformal Prediction Under Adaptive Loggingcited by 0