chrispalmernz/drlnd_p2_continuous_control

Deep Reinforcement Learning Nanodegree - Project 2

Udacity Deep Reinforcement Learning Nanodegree - Project 2

Continuous Control

Use an Agent that teaches an arm to move to a target

Introduction

The movie above is from an Agent I have trained...

The project uses the Unity Reacher environment.

In this environment, a double-jointed arm can move to target locations. A reward of +0.1 is provided for each step that the agent's hand is in the goal location. Thus, the goal of the agent is to maintain its position at the target location for as many time steps as possible.

The observation space consists of 33 variables corresponding to position, rotation, velocity, and angular velocities of the arm. Each action is a vector with four numbers, corresponding to torque applicable to two joints. Every entry in the action vector should be a number between -1 and 1.

Distributed Training

For this project, Udacity provides us with two separate versions of the Unity environment:

• The first version contains a single agent.
• The second version contains 20 identical agents, each with its own copy of the environment.

The second version is useful for algorithms like PPO, A3C, and D4PG that use multiple (non-interacting, parallel) copies of the same agent to distribute the task of gathering experience.

However, the Unity environment has dependencies which must also exist in order for it to run - see Getting Started below ...

Solving the Environment

Note that the project submission needs only solve one of the two versions of the environment, this project solves the second version.

Option 1: Solve the First Version

The task is episodic, and in order to solve the environment, the agent must get an average score of +30 over 100 consecutive episodes.

Option 2: Solve the Second Version

The barrier for solving the second version of the environment is slightly different, to take into account the presence of many agents. In particular, the agents must get an average score of +30 (over 100 consecutive episodes, and over all 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 20 (potentially different) scores. We then take the average of these 20 scores.
• This yields an average score for each episode (where the average is over all 20 agents).

The environment is considered solved, when the average (over 100 episodes) of those average scores is at least +30.

Getting Started

1. Depending somewhat on your operating system, the biggest challenge to getting this going will be that you need to have the Unity ML-Agents libraries and dependencies installed. To support the Udacity Deep Reinforcement Learning Unity envirnonments the specific version 0.40 of Unity ML-Agent is required.

2. There is a guide for installation under Linux and Mac, and a separate guide for Windows users. I am a Windows user, and I found that I needed to do a lot of work to get this going to support my GPU - as a supporting detailed guide points out with great emphasis, it has dependencies on specific versions of Tensorflow, CUDA toolkit and CUDNN. It is highly recommended that this be done in a separate conda environment as the dependencies are likely to be older than current library versions.

3. After installing Unity ML-Agents, download the environment from one of the links below. You need only select the environment that matches your operating system:

   • Version 1: One (1) Agent

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

   • Version 2: Twenty (20) Agents

     • 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 (version 1) or this link (version 2) 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.)

4. Place the file in a working directory such as p2_continuous-control/, and unzip (or decompress) the file.

Instructions

The Jupyter Notebook Continuous_Control.ipynb must be followed to reproduce the training performed. It calls Agent and Model code in the files ddpg_agent.py and model.py. These must also be located in the working directory.

One thing that needs changing is the reacherpath definition early in the notebook - it must contain the path to wherever you have extracted the Reacher environment, as described in the Getting Started section above.

Report

Please see the report.md for a discussion of the algorith and model, and results of running the experiment.