Jake Grigsby

deep_control. Deep Reinforcement Learning for Continuous Control in PyTorch

super_sac. A general model-free off-policy actor-critic implementation. Continuous and Discrete Soft Actor-Critic with multimodal observations, data augmentation, offline learning and behavioral cloning.

AdvancedPacmanDQNs. Extensions of Deep Q Learning

dmc_remastered. A version of the DeepMind Control Suite with randomly generated graphics, for measuring visual generalization in continuous control.

keras-rl. Deep Reinforcement Learning for Keras.

cc-afbc. Advantage-Filtered Behavioral Cloning for Offline Continuous Control

lunarDQfD. example of keras-rl dqfd implementation.

articanon. Neural Text Generation

supersonic. Multiworker PPO with random network distillation in eager execution Tensorflow