Eclectic-Sheep / sheeprl

Distributed Reinforcement Learning accelerated by Lightning Fabric
https://eclecticsheep.ai
Apache License 2.0
302 stars 30 forks source link

Dreamer V1 #26

Closed michele-milesi closed 1 year ago

michele-milesi commented 1 year ago

We would like to implement Dreamer V1, a model-based algorithms that learns long-horizon behaviours by latent imagination. For more information see https://doi.org/10.48550/arXiv.1912.01603.

A WIP can be found here: #27