Jonathan N. Lee's repositories
decision-pretrained-transformer
Implemention of the Decision-Pretrained Transformer (DPT) from the paper Supervised Pretraining Can Learn In-Context Reinforcement Learning.
consumable-irl
We provide the functionality to learn a set of consumable rewards from a set of demonstrations
modular_rl
Implementation of TRPO and related algorithms
policy_gradients
Tensorflow Policy Gradients
zek-control
zek control files for net training