mlsanityData to 7 Oct 2026

Asymptotically Efficient Off-Policy Evaluation for Tabular Reinforcement Learning

arXiv 2001.10742 ↗First posted 29 Jan 2020cs.LG cs.AI stat.MLcited by 60

Not on the map. Why?

Cited by 25 in Offline RL and imitation learning from data, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

We consider the problem of off-policy evaluation for reinforcement learning, where the goal is to estimate the expected reward of a target policy π using offline data collected by running a logging policy μ. Standard importance-sampling based approaches for this problem suffer from a variance that scales exponentially with time horizon H, which motivates a splurge of recent interest in alternatives that break the "Curse of Horizon" (Liu et al. 2018, Xie et al. 2019). In particular, it was shown that a marginalized importance sampling (MIS) approach can be used to achieve an estimation error of order O(H3/ n) in mean square error (MSE) under an episodic Markov Decision Process model with finite states and potentially infinite actions. The MSE bound however is still a factor of H away from a Cramer-Rao lower bound of order Omega(H2/n). In this paper, we prove that with a simple modification to the MIS estimator, we can asymptotically attain the Cramer-Rao lower bound, provided that the action space is finite. We also provide a general method for constructing MIS estimators with high-probability error bounds.

Key facts. Cited by 60 papers in the map (data through 7 Oct 2026). First posted on arXiv 29 Jan 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2020: 3; 2021: 20; 2022: 14; 2023: 7; 2024: 8; 2025: 3; 2026: 5.

60 papers.

Sort:

By area:

  1. 2020Is Pessimism Provably Efficient for Offline RL?cited by 362
  2. 2021Provable Benefits of Actor-Critic Methods for Offline Reinforcement Learningcited by 109
  3. 2022A Review of Off-Policy Evaluation in Reinforcement Learningcited by 92
  4. 2021Towards Instance-Optimal Offline Reinforcement Learning with Pessimismcited by 78
  5. 2022Near-optimal Offline Reinforcement Learning with Linear Representation: Leveraging Variance Information with Pessimismcited by 63
  6. 2021Near-Optimal Offline Reinforcement Learning via Double Variance Reductioncited by 56
  7. 2021Finite Sample Analysis of Minimax Offline Reinforcement Learning: Completeness, Fast Rates and First-Order Efficiencycited by 54
  8. 2021Bootstrapping Fitted Q-Evaluation for Off-Policy Inferencecited by 40
  9. 2021Nearly Horizon-Free Offline Reinforcement Learningcited by 36
  10. 2021Variance-Aware Off-Policy Evaluation with Linear Function Approximationcited by 34
  11. 2021A Minimax Learning Approach to Off-Policy Evaluation in Confounded Partially Observable Markov Decision Processescited by 34
  12. 2021A Statistical Analysis of Polyak-Ruppert Averaged Q-learningcited by 29
38 more
  1. 2021Towards Theoretical Understandings of Robust Markov Decision Processes: Sample Complexity and Asymptoticscited by 27
  2. 2021Mitigating Covariate Shift in Imitation Learning via Offline Data Without Great Coveragecited by 26
  3. 2021Offline Neural Contextual Bandits: Pessimism, Optimization and Generalizationcited by 25
  4. 2020Sparse Feature Selection Makes Batch Reinforcement Learning More Sample Efficientcited by 24
  5. 2021Optimal policy evaluation using kernel-based temporal difference methodscited by 24
  6. 2022Offline Reinforcement Learning with Differential Privacycited by 19
  7. 2022Off-Policy Fitted Q-Evaluation with Differentiable Function Approximators: Z-Estimation and Inference Theorycited by 18
  8. 2021Towards Automatic Evaluation of Dialog Systems: A Model-Free Off-Policy Evaluation Approachcited by 16
  9. 2021Optimal Uniform OPE and Model-based Offline Reinforcement Learning in Time-Homogeneous, Reward-Free and Task-Agnostic Settingscited by 14
  10. 2022Offline Reinforcement Learning with Closed-Form Policy Improvement Operatorscited by 14
  11. 2022On Gap-dependent Bounds for Offline Reinforcement Learningcited by 13
  12. 2024Offline Multi-task Transfer RL with Representational Penalizationcited by 13
  13. 2022Sample Complexity of Nonparametric Off-Policy Evaluation on Low-Dimensional Manifolds using Deep Networkscited by 12
  14. 2022Offline Policy Evaluation and Optimization under Confoundingcited by 12
  15. 2021Non-asymptotic Confidence Intervals of Off-policy Evaluation: Primal and Dual Boundscited by 11
  16. 2023Policy Finetuning in Reinforcement Learning via Design of Experiments using Offline Datacited by 11
  17. 2022Bellman Residual Orthogonalization for Offline Reinforcement Learningcited by 10
  18. 2022Off-policy estimation of linear functionals: Non-asymptotic theory for semi-parametric efficiencycited by 10
  19. 2025Offline Reinforcement Learning in Large State Spaces: Algorithms and Guaranteescited by 10
  20. 2021Minimax Off-Policy Evaluation for Multi-Armed Banditscited by 9
  21. 2020Efficient Evaluation of Natural Stochastic Policies in Offline Reinforcement Learningcited by 8
  22. 2023When is Agnostic Reinforcement Learning Statistically Tractable?cited by 8
  23. 2021Sample Complexity of Offline Reinforcement Learning with Deep ReLU Networkscited by 7
  24. 2022A Complete Characterization of Linear Estimators for Offline Policy Evaluationcited by 7
  25. 2022Scaling Marginalized Importance Sampling to High-Dimensional State-Spaces via State Abstractioncited by 7
  26. 2023Off-policy Evaluation in Doubly Inhomogeneous Environmentscited by 7
  27. 2024An Optimal Tightness Bound for the Simulation Lemmacited by 7
  28. 2023Kernel-based off-policy estimation without overlap: Instance optimality beyond semiparametric efficiencycited by 6
  29. 2024Unraveling the Interplay between Carryover Effects and Reward Autocorrelations in Switchback Experimentscited by 6
  30. 2022Low Variance Off-policy Evaluation with State-based Importance Samplingcited by 5
  31. 2023Model-based Offline Reinforcement Learning with Local Misspecificationcited by 5
  32. 2024Low-Rank Bandits via Tight Two-to-Infinity Singular Subspace Recoverycited by 5
  33. 2022Offline Stochastic Shortest Path: Learning, Evaluation and Towards Optimalitycited by 3
  34. 2021The Curse of Passive Data Collection in Batch Reinforcement Learningcited by 2
  35. 2021Explaining Off-Policy Actor-Critic From A Bias-Variance Perspectivecited by 2
  36. 2024Multiple-policy Evaluation via Density Estimationcited by 2
  37. 2024Off-policy Evaluation with Deeply-abstracted Statescited by 2
  38. 2024Causal Deepsets for Off-policy Evaluation under Spatial or Spatio-temporal Interferencescited by 2

50 most cited of 60 All 60 (JSON)