Intro

Stable Baselines (Docs) is a cleaned up and easier to use version of OpenAI’s baseline Reinforcement Learning algorithms. They support multiple RL algorithms (PPO, DQN, etc) each of which supports some sub-set of features. The docs, however, don’t include a single table where you can see what all the algorithms support in one place. The table below shows them all at a glance, making it easier to decide which algorithms you can or can’t use based on recurrance, continuous actions, multi-processing, etc.

Algorithms