Jrlzoo

An RL codebase in JAX which focuses on efficiency and simplicity. Each baseline algorithm is self-contained in the corresponding folder.

On-policy RL

• PPO

Off-policy RL

• DDPG
• DQN
• SAC
• TD3

Offline RL

• COMBO
• CQL
• IQL
• TD3BC

Fine-tuning Offline RL

• AWAC
• Off2OnRL
• PEX

Meta RL

• HypeRL
• MAML
• MQL
• PEARL

Exploration

• RND

Acknowledgements

• JaxRL
• Flax
• jax-meta-learning