GitHub - katerakelly/oyster: Implementation of Efficient Off-policy Meta-learning via Probabilistic Context Variables (PEARL)

GitHub - katerakelly/oyster: Implementation of Efficient Off-policy Meta-learning via Probabilistic Context Variables (PEARL)

Implementation of Efficient Off-policy Meta-learning via Probabilistic Context Variables (PEARL) - katerakelly/oyster

1 mentions: @svlevine
Date: 2019/04/20 21:48

Referring Tweets

@svlevine The open-source PEARL code has been refactored (by Kate R and Aurick Z) to be much more usable: t.co/rU1eaOVIhc This is our best-performing meta-RL framework, implemented on top of SAC. In case you want to build on it for NeurIPS or compare :) t.co/Ebiwsdg4nH