/amago

a simple and scalable agent for training adaptive policies with sequence-based RL

Primary LanguagePythonMIT LicenseMIT

Issues