Book Image

Hands-On Deep Learning for Games

By : Micheal Lanham
Book Image

Hands-On Deep Learning for Games

By: Micheal Lanham

Overview of this book

The number of applications of deep learning and neural networks has multiplied in the last couple of years. Neural nets has enabled significant breakthroughs in everything from computer vision, voice generation, voice recognition and self-driving cars. Game development is also a key area where these techniques are being applied. This book will give an in depth view of the potential of deep learning and neural networks in game development. We will take a look at the foundations of multi-layer perceptron’s to using convolutional and recurrent networks. In applications from GANs that create music or textures to self-driving cars and chatbots. Then we introduce deep reinforcement learning through the multi-armed bandit problem and other OpenAI Gym environments. As we progress through the book we will gain insights about DRL techniques such as Motivated Reinforcement Learning with Curiosity and Curriculum Learning. We also take a closer look at deep reinforcement learning and in particular the Unity ML-Agents toolkit. By the end of the book, we will look at how to apply DRL and the ML-Agents toolkit to enhance, test and automate your games or simulations. Finally, we will cover your possible next steps and possible areas for future learning.

Related Content you might be interested in

Current Title:

Small book image
Hands-On Deep Learning for Games
Micheal Lanham
Mar, 2019 | 392 pages
Small book image
Learn Unity ML-Agents - Fundamentals of Unity Machine Learning
Micheal Lanham
Jun, 2018 | 204 pages
Small book image
Hands-On Reinforcement Learning for Games
Micheal Lanham
Jan, 2020 | 432 pages
Small book image
Learn ARCore - Fundamentals of Google ARCore
Micheal Lanham
Mar, 2018 | 274 pages
Table of Contents (18 chapters)
Preface
Preface
Who this book is for
What this book covers
To get the most out of this book
Get in touch
Free Chapter
1
Section 1: The Basics
Section 1: The Basics
2
Deep Learning for Games
Deep Learning for Games
The past, present, and future of DL
Neural networks – the foundation
Multilayer perceptron in TF
TensorFlow Basics
Training neural networks with backpropagation
Building an autoencoder with Keras
Exercises
Summary
3
Convolutional and Recurrent Networks
Convolutional and Recurrent Networks
Convolutional neural networks
Understanding convolution
Building a self-driving CNN
Memory and recurrent networks
Playing Rock, Paper, Scissors with LSTMs
Exercises
Summary
4
GAN for Games
GAN for Games
Introducing GANs
Coding a GAN in Keras
Wasserstein GAN
Generating textures with a GAN
A GAN for creating music
Exercises
Summary
5
Building a Deep Learning Gaming Chatbot
Building a Deep Learning Gaming Chatbot
Neural conversational agents
Sequence-to-sequence learning
DeepPavlov
Building the chatbot server
Running the chatbot in Unity
Exercises
Summary
6
Section 2: Deep Reinforcement Learning
Section 2: Deep Reinforcement Learning
7
Introducing DRL
Introducing DRL
Reinforcement learning
RL with the OpenAI Gym
A Q-Learning model
First DRL with Deep Q-learning
RL experiments
Exercises
Summary
8
Unity ML-Agents
Unity ML-Agents
Installing ML-Agents
Training an agent
What's in a brain?
Monitoring training with TensorBoard
Running an agent
Exercises
Summary
9
Agent and the Environment
Agent and the Environment
Exploring the training environment
Understanding state
Understanding visual state
Convolution and visual state
Recurrent networks for remembering series
Exercises
Summary
10
Understanding PPO
Understanding PPO
Marathon RL
The partially observable Markov decision process
Actor-Critic and continuous action spaces
Understanding TRPO and PPO
Learning to tune PPO
Exercises
Summary
11
Rewards and Reinforcement Learning
Rewards and Reinforcement Learning
Rewards and reward functions
Sparsity of rewards
Curriculum Learning
Understanding Backplay
Curiosity Learning
Exercises
Summary
12
Imitation and Transfer Learning
Imitation and Transfer Learning
IL, or behavioral cloning
Online training
Offline training
Transfer learning
Imitation Transfer Learning
Exercises
Summary
13
Building Multi-Agent Environments
Building Multi-Agent Environments
Adversarial and cooperative self-play
Adversarial self-play
Multi-brain play
Adding individuality with intrinsic rewards
Extrinsic rewards for individuality
Exercises
Summary
14
Section 3: Building Games
Section 3: Building Games
15
Debugging/Testing a Game with DRL
Debugging/Testing a Game with DRL
Introducing the game
Setting up ML-Agents
Overriding the Unity input system
Testing through imitation
Analyzing the testing process
Exercises
Summary
16
Obstacle Tower Challenge and Beyond
Obstacle Tower Challenge and Beyond
The Unity Obstacle Tower Challenge
Deep Learning for your game?
Building your game
More foundations of learning
Summary
17
Other Books You May Enjoy
Other Books You May Enjoy
Leave a review - let other readers know what you think

The partially observable Markov decision process

Back in Chapter 5, Introducing DRL, we learned that a Markov Decision Process (MDP) is used to define the state/model an agent uses to calculate an action/value from. In the case of Q-learning, we have seen how a table or grid could be used to hold an entire MDP for an environment such as the Frozen Pond or GridWorld. These types of RL are model-based, meaning they completely model every state in the environment—every square in a grid game, for instance. Except, in most complex games and environments, being able to map physical or visual state becomes a partially observable problem, or what we may refer to as a partially observable Markov decision process (POMDP).

A POMDP defines a process where an agent never has a complete view of its environment, but instead learns to conduct actions based on a derived general policy....

Previous Section
End of Section 3
Next Section