SAC+

Soft Actor-Critic implementation with SOTA model-free extension (REDQ) and SOTA model-based extension (MBPO).

Algorithms implemented:

• Soft Actor-Critic (SAC)
• Model-Based Policy Optimization (MBPO)
• Randomized Ensembled Double Q-Learning (REDQ)

Acknowledgements

I was inspired by the following repositories:

• https://github.com/vwxyzjn/cleanrl
• https://github.com/DLR-RM/stable-baselines3
• https://github.com/kzl/lifelong_rl