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

Showing 231240 of 265 papers

TitleStatusHype
Off-policy Evaluation with Deeply-abstracted StatesCode0
From Importance Sampling to Doubly Robust Policy GradientCode0
Future-Dependent Value-Based Off-Policy Evaluation in POMDPsCode0
Off-Policy Evaluation for Action-Dependent Non-Stationary EnvironmentsCode0
On (Normalised) Discounted Cumulative Gain as an Off-Policy Evaluation Metric for Top-n RecommendationCode0
Hallucinated Adversarial Control for Conservative Offline Policy EvaluationCode0
Harnessing Distribution Ratio Estimators for Learning Agents with Quality and DiversityCode0
Hindsight-DICE: Stable Credit Assignment for Deep Reinforcement LearningCode0
When is Off-Policy Evaluation (Reward Modeling) Useful in Contextual Bandits? A Data-Centric PerspectiveCode0
Human Choice Prediction in Language-based Persuasion Games: Simulation-based Off-Policy EvaluationCode0
Show:102550
← PrevPage 24 of 27Next →

No leaderboard results yet.