This repository contains my codes, report and other files for Udacity Deep Reinforcement Learning Nanodegree Project 3: Collaboration and Competition.
In this project, 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 environment is based on Unity ML-agents. Unity ML-Agents is an open-source Unity plugin that enables games and simulations to serve as environments for training intelligent agents.
Note: The Unity ML-Agent team frequently releases updated versions of their environment. We are using the v0.4 interface. The project environment provided by Udacity is similar to, but not identical to the TennisEnv environment on the Unity ML-Agents GitHub page.
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, the 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.
In this project I used a multi-agent algorithm called Multi Agent Deep Deterministic Policy Gradient (MADDPG) which is described in the paper Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments.
-
To begin with, you need to configure a Python 3.6 / PyTorch 0.4.0 environment with the requirements described in Udacity repository
-
Then you need to clone this project and have it accessible in your Python environment
-
For this project, you will not need to install Unity. This is because we have already built the environment for you, and you can download it from one of the links below. You need to 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.)
-
Finally, unzip the environment archive in the 'project's environment' directory and eventually adjust the path to the UnityEnvironment in the code.
You can either run Tennis.ipynb in the Udacity Online Workspace for "Project3: Collaboration and Competition" step by step or build your own local environment and set the path to the UnityEnvironment in the code.
Note: The Workspace does not allow you to see the simulator of the environment; so, if you want to watch the agent while it is training, you should train locally.