The idea behind this reposity is to build Reinforcement Learning solutions to different type of games / environments.

• python 3.7

• PPO
  • BipedalWalker-v2
    python ppo.py -env BipedalWalker-v3
  • LunarLanderContinuous
    python ppo.py -env LunarLanderContinuous-v2 -e 50

• DDPG
  • MontainCarContinuous
    python ddpg.py -env MountainCarContinuous-v0 -t 100 -noise 1
  • Pendulum-v0
    python ddpg.py -env Pendulum-v0 -t 200

• REINFORCE

  • Cartpole

• Q-Learning

  • TicTacToe

• DQN

  • Cartpole
  • Pong
  • SpaceInvaders

• Breakout
  • Solution done with Samuel Heinrichs - Link to video in image below