Yilingeacc / reinforcement-learning

reinforcement learning

reinforcement learning algorithms implementation with pytorch

reference:

https://github.com/AI4Finance-LLC/ElegantRL (DRL algo impl)

https://github.com/starry-sky6688/StarCraft (MARL algo impl)

content

1.Tabular:

• MazeEnv: my gmy-like environment, for tabular algos

• MonteCarlo

• off-policy MonteCarlo (with important sampling)

• Sarsa

• QLearning

• DoubleQLearning

• n-step Sarsa

• Sarsa(lambda)

2.Deep Q Network

• Deep Q Network

• DDQN(Double DQN)

• Dueling DQN

• D3QN(Dueling DDQN)

3.Policy Gradient

• REINFORCE

• REINFORCE with baseline

4.Deterministic Policy Gradient

• DDPG(Deep Deterministic Policy Gradient)

• TD3(Twin Delayed DDPG)

5.Actor Critic

• PPO(Proximal Policy Optimization) (PPO-Clip)

6.Multi-Agent RL

(Env: Multi-Agent partical world)

• Qmix