pieromacaluso/collaboration-n-competition

Implementation of Multi-Agent Deep Deterministic Policy Gradient (MADDPG) for the Tennis environment in the context of "Collaboration and Competition", the third Udacity Deep Reinforcement Learning Nanodegree project.

Project 3: Collaboration'n'Competition

Introduction

For this project, we worked with the Tennis environment.

In this environment, two agents control rackets to bounce a ball over a net. 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.

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

The task is episodic, and in order to solve the environment, 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.

Getting Started

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 root directory of this GitHub repository and unzip (or decompress) the file.

3. Two alternatives there:

   • Poetry: Run poetry install in the root directory.
   • Other dependency management tools: Install the requirements listed in the requirements.txt file.

How-to start the training of the agent

To train the agent it is necessary to go to the root folder of the project and run python train.py.

The code will start to create checkpoints in the appropriate folder.

How-to test the agent

The best parameters of the network found in the training are available in the file checkpoint_a0.pth and checkpoint_a1.pth in the root folder of this repository, one for each agent in the environment.

In order to test the network it is possible to go to the root folder and run python test.py

Result Show Case