Saoussen-CH / DRL_Navigation

Deep Reinforcement Learning, Value Based Method, Deep Q-Learning, PyTorch

Project 1: Navigation

Overview

This project is solution for the first prject of the Udacity's Deep Reinforcement Learning Nanodegree (Facebook PyTorch Nanodegree Scholarship Phase 3). Value-Based-Methods such as Deep Q-Network (DQN) is used to train an agent to naviagte the world and collect yellow bananas while avoiding blue bananas.

Introduction

For this project, you will train an agent to navigate (and collect bananas!) in a large, square world.

A reward of +1 is provided for collecting a yellow banana, and a reward of -1 is provided for collecting a blue banana. Thus, the goal of your agent is to collect as many yellow bananas as possible while avoiding blue bananas.

The state space has 37 dimensions and contains the agent's velocity, along with ray-based perception of objects around agent's forward direction. Given this information, the agent has to learn how to best select actions. Four discrete actions are available, corresponding to:

• 0 - move forward.
• 1 - move backward.
• 2 - turn left.
• 3 - turn right.

The task is episodic, and in order to solve the environment, your agent must get an average score of +13 over 100 consecutive episodes.

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 environment.

2. Then, download the folder of this DRL_Navigation GitHub repository, and unzip (or decompress) the file and change the path in the argument for creating the environment under the notebook Navigation_solution.ipynb:

 env = UnityEnvironment(file_name=" Your Path")

Description

• dqn_agent.py: code for the agent used in the environment
• model.py: code containing the Deep Q-Network architecture used as the function approximator by the agent
• checkpoint_v1.pth: saved model weights for the Deep Q-Network model
• Navigation.ipynb: notebook containing the Project 1 Navigation Solution

Instructions

Follow the instructions in Navigation_solution.ipynb to get started with training your own agent! To watch a trained smart agent, use the checkpoint checkpoint_v1.pth for loading the trained model.

Enhancements

Several enhancements to the original DQN algorithm have also been incorporated:

• Double DQN [Paper]
• Prioritized Experience Replay [Paper]
• Dueling DQN [Paper]

Results

Plot showing the score per episode over all the episodes. The environment was solved in 402 episodes.

Deep Q-Network