/Discrete_IVR

Discrete version of SQL in "Offline RL with No OOD Actions: In-Sample Learning via Implicit Value Regularization"

Primary LanguagePythonMIT LicenseMIT

Discrete version of "Offline RL with No OOD Actions: In-Sample Learning via Implicit Value Regularization"

This repository implements the discrete version of BCQ, IQL, CQL and SQL based on d3rlpy.