Deep-rl-mxnet

Mxnet implementation of Deep Reinforcement Learning papers.

Now this repository contains:

1. DQN [code]

• Playing Atari with Deep Reinforcement Learning
• Human-level control through deep reinforcement learning

2. Double DQN [code]

• Deep Reinforcement Learning with Double Q-learning

3. Dueling DQN [code]

• Dueling Network Architectures for Deep Reinforcement Learning

4. Policy Gradient [code]

• Policy Gradient Methods for Reinforcement Learning with Function Approximation

5. Deep Deterministic Policy Gradient [code] (Detailed implementation) :star:

• Continuous Control with Deep Reinforcement Learning

6. Proximal Policy Optimization [code]

• Proximal Policy Optimization Algorithms
• Emergence of Locomotion Behaviours in Rich Environments

7. TD3 [code] (Very detailed implementation) :star: :star:

• Addressing Function Approximation Error in Actor-Critic Methods

8. A2C [code]

• Asynchronous Methods for Deep Reinforcement Learning

Installation

$ git clone https://github.com/ZhengXinyue/Deep-rl-mxnet
$ cd Deep-rl-mxnet

create & activate virtual env then install dependency:

with venv/virtualenv + pip:

$ python -m venv env  # use `virtualenv env` for Python2, use `python3 ...` for Python3 on Linux & macOS
$ source env/bin/activate  # use `env\Scripts\activate` on Windows
$ pip install -r requirements.txt

If you get something like this:

unable to execute 'swig': No such file or directory

do:

sudo apt-get install swig

Mujoco Installation(Optional):

Please refer to this repository

Examples:

Maybe In the future:

9. SAC

• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor