To be concise, This repository is a solution to reference instructions.pdf We create an environment(a game) and different RL (Reinforcement Learning) agents that can interact (play). The following are the three RL agents that have been implemented.
- Monte Carlo
- Backward-view Sarsa lambda
- Backward-view Sarsa lambda using a linear function approximator
A human being can also play the game with or without the assist of any of the agents.