Learn Unity ML-Agents：Fundamentals of Unity Machine Learning

封面

Title Page

Copyright and Credits

Learn Unity ML - Agents - Fundamentals of Unity Machine Learning

Dedication

Packt Upsell

Why subscribe?

PacktPub.com

Contributors

About the author

About the reviewers

Packt is searching for authors like you

Preface

Who this book is for

What this book covers

To get the most out of this book

Download the example code files

Download the color images

Conventions used

Get in touch

Reviews

Introducing Machine Learning and ML-Agents

Machine Learning

Training models

A Machine Learning example

ML uses in gaming

ML-Agents

Running a sample

Setting the agent Brain

Creating an environment

Renaming the scripts

Academy Agent and Brain

Setting up the Academy

Setting up the Agent

Setting up the Brain

Exercises

Summary

The Bandit and Reinforcement Learning

Reinforcement Learning

Configuring the Agent

Contextual bandits and state

Building the contextual bandits

Creating the ContextualDecision script

Updating the Agent

Exploration and exploitation

Making decisions with SimpleDecision

MDP and the Bellman equation

Q-Learning and connected agents

Looking at the Q-Learning ConnectedDecision script

Exercises

Summary

Deep Reinforcement Learning with Python

Installing Python and tools

Installation

Mac/Linux installation

Windows installation

Docker installation

GPU installation

Testing the install

ML-Agents external brains

Running the environment

Neural network foundations

But what does it do?

Deep Q-learning

Building the deep network

Training the model

Exploring the tensor

Proximal policy optimization

Implementing PPO

Understanding training statistics with TensorBoard

Exercises

Summary

Going Deeper with Deep Learning

Agent training problems

When training goes wrong

Fixing sparse rewards

Fixing the observation of state

Convolutional neural networks

Experience replay

Building on experience

Partial observability memory and recurrent networks

Partial observability

Memory and recurrent networks

Asynchronous actor – critic training

Multiple asynchronous agent training

Exercises

Summary

Playing the Game

Multi-agent environments

Adversarial self-play

Using internal brains

Using trained brains internally

Decisions and On-Demand Decision Making

The Bouncing Banana

Imitation learning

Setting up a cloning behavior trainer

Curriculum Learning

Exercises

Summary

Terrarium Revisited – A Multi-Agent Ecosystem