wjlgatech / drl-marl-tennis

Playing Tennis: An Example of Multi-Agent Reinforcement Learning

Objective

To train 2 agents to play tennis.

Overview

Environment: In this environment, two agents control rackets to bounce a ball over a net. Thus, the goal of each agent is to keep the ball in play.

NOTE: This environment is a modified version of Unity ML Tennis. Do NOT use the Unity version.

Observation Space: The observation space consists of 8 variables corresponding to the position and velocity of the ball and racket. Each agent receives its own, local observation.

Action Space: Two continuous actions are available, corresponding to movement toward (or away from) the net, and jumping.

Reward: If an agent hits the ball over the net, it receives a reward of +0.1. If an agent lets a ball hit the ground or hits the ball out of bounds, it receives a reward of -0.01. Thus, the goal of each agent is to keep the ball in play.

Problem Sovled: The task is episodic, and in order to solve the environment, your agents must get an average score of +0.5 (over 100 consecutive episodes, after taking the maximum over both agents). Specifically,

• After each episode, we add up the rewards that each agent received (without discounting), to get a score for each agent. This yields 2 (potentially different) scores. We then take the maximum of these 2 scores.
• This yields a single score for each episode.

The environment is considered solved, when the average (over 100 episodes) of those scores is at least +0.5.

Instructions

STEP 1: Activate Dependencies

Follow the instructions in the DRLND github repo to set up your python environment.

1. Create (and activate) a new environment with Python 3.6.

   • Linux or Mac:

   conda create --name drlnd python=3.6
   source activate drlnd

   • Windows:

   conda create --name drlnd python=3.6 
   activate drlnd

2. Follow the instructions in this repository to perform a minimal install of OpenAI gym.

   • Next, install the classic control environment group by following the instructions here.
   • Then, install the box2d environment group by following the instructions here.

3. Clone the repository (if you haven't already!), and navigate to the python/ folder. Then, install several dependencies.

git clone https://github.com/udacity/deep-reinforcement-learning.git
cd deep-reinforcement-learning/python
pip install .

4. Create an IPython kernel for the drlnd environment.

python -m ipykernel install --user --name drlnd --display-name "drlnd"

5. Before running code in a notebook, change the kernel to match the drlnd environment by using the drop-down Kernel menu.

STEP 2: Clone this Repository

Open up a terminal, go to the directory of your choice and clone the repository

git clone https://github.com/wjlgatech/DRL-marl-tennis.git .

STEP 3: Download the Environment

1. Download the environment from one of the links below. You need only select the environment that matches your operating system:

   • Linux: click here
   • Mac OSX: click here
   • Windows (32-bit): click here
   • Windows (64-bit): click here

   (For Windows users) Check out this link if you need help with determining if your computer is running a 32-bit version or 64-bit version of the Windows operating system.

   (For AWS) If you'd like to train the agent on AWS (and have not enabled a virtual screen), then please use this link to obtain the "headless" version of the environment. You will not be able to watch the agent without enabling a virtual screen, but you will be able to train the agent. (To watch the agent, you should follow the instructions to enable a virtual screen, and then download the environment for the Linux operating system above.)

2. Place the file in the folder of the above local repository, and unzip (or decompress) the file.

STEP 4: Run the codes

One way is that you open the Tennis.ipynb notebook to follow instructions there:

jupyter notebook Tennis.ipynb

Another way is that you train the agents with

python train.py

and test the trained agents with

python test.py

Code Overview

The code consists of the following modules

Tennis.ipynb - the main notebook
Report.ipynb - the report of this project
maddpg_agent.py - defines the Agent that is to be trained
maddpg_model.py - defines the MADDPG model for the Actor and the Critic network
checkpoint_actor1.pth - is the final trained Actor network
checkpoint_criti1c.pth - is the final trained Critic network
train.py - train the MADDPG agent
test.py - test the performance of the trained agent

Results

Environment solved in 5305 episodes with Average Score 0.505 (>=0.5).