Deep RL Vectorized Environment

[GeorgeS2019]

Current Status

RL Environments

• K-armed bandit

  In K-armed bandit RL Environment, an agent has to choose between K different actions (or arms) in order to maximize its reward. The goal of the K-armed bandit problem is to find the optimal policy for an agent to take in an environment.

RL Space

SB3 RL Environments comparisons

The RL Environment come from gymnasium the replacement of OpenAI Gym

The great Test-Driven Development coverage of the RL. Environment can help you design not from scratch but well thought RL software architect Design patterns that are well tested.

VecEnvWrapper or Vectorized Environments Wrappers