An index of algorithms for offline reinforcement learning (offline-rl)

Open Bandit Pipeline: a python library for bandit algorithms and off-policy evaluation

Implementations and examples of common offline policy evaluation methods in Python.

SCOPE-RL: A python library for offline reinforcement learning, off-policy evaluation, and selection

Reinforcement Learning Short Course

(WSDM2022 Best Paper Award Runner-Up) "Doubly Robust Off-Policy Evaluation for Ranking Policies under the Cascade Behavior Model"

Representation Learning for OPE

(KDD2023) "Off-Policy Evaluation of Ranking Policies under Diverse User Behavior"

(NeurIPS2023) "Future-Dependent Value-Based Off-Policy Evaluation in POMDPs"

Off-Policy Interval Estimation withConfounded Markov Decision Process

Robust Offline Reinforcement Learning with Heavy-Tailed Rewards

Implementation of "Deeply-Debiased Off-Policy Interval Estimation" (ICML, 2021) in Python

Stateful implementations of OPE algorithms, designed for use in the development of offline RL models

Official implementation for "On the Reuse Bias in Off-Policy Reinforcement Learning" (IJCAI 2023)

Omitting-States-Irrelevant-to-Return Importance Sampling estimator for off-policy evaluation

Implementation of Deep Jump Learning for Off-Policy Evaluation in Continuous Treatment Settings (NeurIPS, 2021) in Python

[NeurIPS 2023] Counterfactual-Augmented Importance Sampling for Semi-Offline Policy Evaluation. https://arxiv.org/abs/2310.17146

Conformal Off-policy Prediction

Implementation of "A Minimax Learning Approach to Off-Policy Evaluation in Confounded Partially Observable Markov Decision Processes" (ICML)

Implementation of "Off-Policy Interval Estimation with Confounded Markov Decision Process" (JASA, 2022+)