SOTAVerified

Off-policy evaluation

Off-policy Evaluation (OPE), or offline evaluation in general, evaluates the performance of hypothetical policies leveraging only offline log data. It is particularly useful in applications where the online interaction involves high stakes and expensive setting such as precision medicine and recommender systems.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 151–200 of 265 papers

TitleStatusHype
Off-Policy Evaluation in Partially Observed Markov Decision Processes under Sequential Ignorability—0
Stateful Offline Contextual Policy Evaluation and Learning—0
Why Should I Trust You, Bellman? Evaluating the Bellman Objective with Off-Policy Data—0
A Spectral Approach to Off-Policy Evaluation for POMDPs—0
Data-Driven Off-Policy Estimator Selection: An Application in User Marketing on An Online Content Delivery Service—0
Accelerating Offline Reinforcement Learning Application in Real-Time Bidding and Recommendation: Potential Use of Simulation—0
State Relevance for Off-Policy EvaluationCode0
Evaluating the Robustness of Off-Policy EvaluationCode1
Debiasing Samples from Online Learning Using Bootstrap—0
Model Selection for Offline Reinforcement Learning: Practical Considerations for Healthcare SettingsCode1
Online Learning for Recommendations at Grubhub—0
A Unified Off-Policy Evaluation Approach for General Value Function—0
Supervised Off-Policy RankingCode0
Unifying Gradient Estimators for Meta-Reinforcement Learning via Off-Policy EvaluationCode1
Variance-Aware Off-Policy Evaluation with Linear Function Approximation—0
Active Offline Policy SelectionCode1
Offline RL Without Off-Policy EvaluationCode1
Control Variates for Slate Off-Policy EvaluationCode0
A Deep Reinforcement Learning Approach to Marginalized Importance Sampling with the Successor RepresentationCode1
Robust Generalization despite Distribution Shift via Minimum Discriminating InformationCode0
Deep Proxy Causal Learning and its Application to Confounded Bandit Policy EvaluationCode0
Off-Policy Evaluation via Adaptive Weighting with Data from Contextual BanditsCode1
Deeply-Debiased Off-Policy Interval EstimationCode0
Autoregressive Dynamics Models for Offline Policy Evaluation and Optimization—0
Universal Off-Policy EvaluationCode0
Discovering an Aid Policy to Minimize Student Evasion Using Offline Reinforcement Learning—0
Off-Policy Risk Assessment in Contextual Bandits—0
Benchmarks for Deep Off-Policy EvaluationCode1
Infinite-Horizon Offline Reinforcement Learning with Linear Function Approximation: Curse of Dimensionality and Algorithm—0
Non-asymptotic Confidence Intervals of Off-policy Evaluation: Primal and Dual Bounds—0
Minimax Model Learning—0
Towards Automatic Evaluation of Dialog Systems: A Model-Free Off-Policy Evaluation ApproachCode0
Off-policy Confidence Sequences—0
Bootstrapping Fitted Q-Evaluation for Off-Policy Inference—0
Finite Sample Analysis of Minimax Offline Reinforcement Learning: Completeness, Fast Rates and First-Order Efficiency—0
Minimax Off-Policy Evaluation for Multi-Armed Bandits—0
Smoothed functional-based gradient algorithms for off-policy reinforcement learning: A non-asymptotic viewpoint—0
Off-Policy Evaluation of Slate Policies under Bayes Risk—0
Practical Marginalized Importance Sampling with the Successor Representation—0
Optimal Mixture Weights for Off-Policy Evaluation with Multiple Behavior Policies—0
Reliable Off-policy Evaluation for Reinforcement Learning—0
Harnessing Distribution Ratio Estimators for Learning Agents with Quality and DiversityCode0
Deep Jump Learning for Off-Policy Evaluation in Continuous Treatment SettingsCode0
Off-Policy Interval Estimation with Lipschitz Value Iteration—0
Off-Policy Evaluation of Bandit Algorithm from Dependent Samples under Batch Update Policy—0
A Practical Guide of Off-Policy Evaluation for Bandit Problems—0
CoinDICE: Off-Policy Confidence Interval Estimation—0
Optimal Off-Policy Evaluation from Multiple Logging PoliciesCode1
Deep Jump Q-Evaluation for Offline Policy Evaluation in Continuous Action Space—0
Open Bandit Dataset and Pipeline: Towards Realistic and Reproducible Off-Policy EvaluationCode1
Show:102550
← PrevPage 4 of 6Next →

No leaderboard results yet.