Pith. sign in

REVIEW 18 cited by

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2212.06355 v1 pith:UFN3Y4ON submitted 2022-12-13 stat.ML cs.LGmath.STstat.MEstat.TH

A Review of Off-Policy Evaluation in Reinforcement Learning

classification stat.ML cs.LGmath.STstat.MEstat.TH
keywords learningbeenevaluationmethodsnumberoff-policyreinforcementresearch
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Reinforcement learning (RL) is one of the most vibrant research frontiers in machine learning and has been recently applied to solve a number of challenging problems. In this paper, we primarily focus on off-policy evaluation (OPE), one of the most fundamental topics in RL. In recent years, a number of OPE methods have been developed in the statistics and computer science literature. We provide a discussion on the efficiency bound of OPE, some of the existing state-of-the-art OPE methods, their statistical properties and some other related research directions that are currently actively explored.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 18 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Estimating Causal Effects from Data Generated by Stochastic Algorithms

    stat.ME 2026-07 accept novelty 8.0

    Logging the features and relative probability of one unexposed item alongside the exposed item identifies causal effects of content features from stochastic algorithms even with unobserved confounders.

  2. Fitted Occupancy-Ratio Evaluation without Bellman Completeness

    stat.ML 2026-07 accept novelty 7.0

    FORE estimates discounted occupancy ratios by iterating KL-projected adjoint Bellman updates, achieving convergence under ratio realizability alone without Bellman completeness.

  3. Off-Policy Evaluation for Missingness-Aware Policies in MDPs with Rewards Missing Not at Random

    stat.ML 2026-06 unverdicted novelty 7.0

    Identifies full-data conditional mean rewards under MNAR missingness via shadow variables and a bridge function, then builds a consistent FQE-style OPE estimator for missingness-aware policies.

  4. Off-Policy Evaluation with Strategic Agents via Local Disclosure

    cs.AI 2026-06 unverdicted novelty 7.0

    Local disclosure via post-hoc explanations enables consistent doubly robust estimation of policy value in one-shot OPE with strategic agents by recovering pre-strategic covariates under a conditional log-normal cost s...

  5. Bandit Simulation for Average Reward Inference

    stat.ML 2026-05 unverdicted novelty 7.0

    BSI fits an environment simulator from bandit data and propagates parameter uncertainty to produce asymptotically valid confidence intervals for mean reward under arbitrary evaluation policies, including black-box ada...

  6. Offline Contextual Bandits in the Presence of New Actions

    cs.LG 2026-05 unverdicted novelty 7.0

    PONA integrates the LCPI estimator for new action selection with the DR estimator for existing actions to optimize policies in offline contextual bandits with evolving action spaces.

  7. Logging Policy Design for Off-Policy Evaluation

    stat.ML 2026-05 unverdicted novelty 7.0

    Derives optimal logging policies for off-policy evaluation by balancing reward concentration against action coverage in known, unknown, and partially known regimes of target policy and rewards.

  8. Off-Policy Learning with Limited Supply

    cs.LG 2026-03 unverdicted novelty 7.0

    OPLS is a new off-policy method for contextual bandits with limited supply that outperforms greedy approaches by prioritizing items with higher relative expected rewards for the current user.

  9. Anytime-valid Optimal Policy Identification

    stat.ME 2026-06 unverdicted novelty 6.0

    Constructs a time-indexed set S_t retaining the true optimal policy uniformly over time with high probability, enabling early stopping with sample complexity O((log |Π| + log log(1/Δ_min))/Δ_min²) when the optimum is unique.

  10. Precision Physical Activity Prescription via Reinforcement Learning for Functional Actions

    stat.AP 2026-05 unverdicted novelty 6.0

    Develops an offline RL algorithm for functional actions that learns personalized optimal daily step distributions from All of Us data to lower cardiometabolic risk.

  11. Precision Physical Activity Prescription via Reinforcement Learning for Functional Actions

    stat.AP 2026-05 unverdicted novelty 6.0

    Develops a new offline RL algorithm for functional actions to learn optimal personalized daily step distributions from All of Us observational data that reduce cardiometabolic risk.

  12. An adaptive variance estimator for relative sparsity

    stat.ME 2026-05 unverdicted novelty 6.0

    A new adaptive variance estimator for relative sparsity coefficients is introduced that fully utilizes the prior asymptotic normality theorem and incorporates variable selection effects.

  13. Distributional Off-Policy Evaluation with Deep Quantile Process Regression

    stat.ML 2026-04 unverdicted novelty 6.0

    DQPOPE estimates the entire return distribution in off-policy evaluation via deep quantile process regression, providing statistical advantages over standard single-value methods with equivalent sample sizes.

  14. Off-Policy Learning with Limited Supply

    cs.LG 2026-03 unverdicted novelty 6.0

    OPLS is a new off-policy learning method for contextual bandits with limited supply that outperforms conventional greedy approaches by prioritizing items with relatively higher expected rewards compared to other users.

  15. Fed-CausalDiff: Decoupled Synchronization for Federated Do-Simulation and Policy Evaluation

    cs.LG 2026-06 unverdicted novelty 5.0

    Fed-CausalDiff proposes decoupled synchronization in a federated causal diffusion model to improve do-simulation and policy-value estimation across heterogeneous decentralized datasets.

  16. Counterfactually Safe Reinforcement Learning

    stat.ML 2026-05 unverdicted novelty 5.0

    Formalizes counterfactual individual harm in RL and introduces a two-stage policy learning method with finite-sample guarantees on sub-optimality gap and harm rate control.

  17. Logging Policy Design for Off-Policy Evaluation

    stat.ML 2026-05 unverdicted novelty 5.0

    Derives optimal logging policies for minimizing off-policy evaluation error under known, unknown, and partially known target policies and reward distributions.

  18. A Review of Causal Decision Making

    stat.ML 2025-02 unverdicted novelty 2.0

    A review that organizes causal decision making into three stages and consolidates methods into an open Python collection.