razrhj / rl-algorithms-by-pytorch

Clean and robust implementations of Reinforcement Learning algorithms by Pytorch

RL-Algorithms-by-Pytorch

$\color{red}{Click\ \ the\ \ name\ \ of\ \ Algorithm\ (in\ \ blue)\ to\ \ enter\ \ the\ \ respective\ \ repositories\ !}$

I found the current implementations of Reinforcement Learning Algorithms are somewhat complicated, which is hard to get start.

Here are some classical Reinforcement Learning Algorithms implemented by Pytorch. I tried to make them clean, robust, and unified, hoping to help you get start with RL quickly.

Now I have finished Q-learning, DQN, DDQN, PPO discrete, PPO continuous, TD3, SAC Continuous, SAC Discrete. I will implement more in the future.

Q-learning:

DQN/DDQN on Classic Control:

DQN/DDQN on Atari Game:

Pong	Enduro

PPO Discrete:

PPO Continuous:

TD3:

SAC Continuous:

SAC Discrete:

References

DQN: Mnih V , Kavukcuoglu K , Silver D , et al. Playing Atari with Deep Reinforcement Learning[J]. Computer Science, 2013.

Double DQN: Hasselt H V , Guez A , Silver D . Deep Reinforcement Learning with Double Q-learning[J]. Computer ence, 2015.

PPO: Proximal Policy Optimization Algorithms, Emergence of Locomotion Behaviours in Rich Environments

TD3: Fujimoto S , Hoof H V , Meger D . Addressing Function Approximation Error in Actor-Critic Methods[J]. 2018.

SAC:

Haarnoja T, Zhou A, Abbeel P, et al. Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor[C]//International conference on machine learning. PMLR, 2018: 1861-1870.

Soft Actor-Critic Algorithms and Applications

Christodoulou P. Soft actor-critic for discrete action settings[J]. arXiv preprint arXiv:1910.07207, 2019.