Hands-On Deep Learning for Games电子书

Dedication

About Packt

Why subscribe?

Packt.com

Contributors

About the author

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

Section 1: The Basics

Deep Learning for Games

The past, present, and future of DL

The past

The present

The future

Neural networks – the foundation

Training a perceptron in Python

Multilayer perceptron in TF

TensorFlow Basics

Training neural networks with backpropagation

The Cost function

Partial differentiation and the chain rule

Building an autoencoder with Keras

Training the model

Examining the output

Exercises

Summary

Convolutional and Recurrent Networks

Convolutional neural networks

Monitoring training with TensorBoard

Understanding convolution

Building a self-driving CNN

Spatial convolution and pooling

The need for Dropout

Memory and recurrent networks

Vanishing and exploding gradients rescued by LSTM

Playing Rock, Paper, Scissors with LSTMs

Exercises

Summary

GAN for Games

Introducing GANs

Coding a GAN in Keras

Training a GAN

Optimizers

Wasserstein GAN

Generating textures with a GAN

Batch normalization

Leaky and other ReLUs

A GAN for creating music

Training the music GAN

Generating music via an alternative GAN

Exercises

Summary

Building a Deep Learning Gaming Chatbot

Neural conversational agents

General conversational models

Sequence-to-sequence learning

Breaking down the code

Thought vectors

DeepPavlov

Building the chatbot server

Message hubs (RabbitMQ)

Managing RabbitMQ

Sending and receiving to/from the MQ

Writing the message queue chatbot

Running the chatbot in Unity

Installing AMQP for Unity

Exercises

Summary

Section 2: Deep Reinforcement Learning

Introducing DRL

Reinforcement learning

The multi-armed bandit

Contextual bandits

RL with the OpenAI Gym

A Q-Learning model

Markov decision process and the Bellman equation

Q-learning

Q-learning and exploration

First DRL with Deep Q-learning

RL experiments

Keras RL

Exercises

Summary

Unity ML-Agents

Installing ML-Agents

Training an agent

What's in a brain?

Monitoring training with TensorBoard

Running an agent

Loading a trained brain

Exercises

Summary

Agent and the Environment

Exploring the training environment

Training the agent visually

Reverting to the basics

Understanding state

Understanding visual state

Convolution and visual state

To pool or not to pool

Recurrent networks for remembering series

Tuning recurrent hyperparameters

Exercises

Summary

Understanding PPO

Marathon RL

The partially observable Markov decision process

Actor-Critic and continuous action spaces

Expanding network architecture

Understanding TRPO and PPO

Generalized advantage estimate

Learning to tune PPO

Coding changes required for control projects

Multiple agent policy

Exercises

Summary

Rewards and Reinforcement Learning

Rewards and reward functions

Building reward functions

Sparsity of rewards

Curriculum Learning

Understanding Backplay

Implementing Backplay through Curriculum Learning

Curiosity Learning

The Curiosity Intrinsic module in action

Trying ICM on Hallway/VisualHallway

Exercises

Summary

Imitation and Transfer Learning

IL, or behavioral cloning

Online training

Offline training

Setting up for training

Feeding the agent

Transfer learning

Transferring a brain

Exploring TensorFlow checkpoints

Imitation Transfer Learning

Training multiple agents with one demonstration

Exercises

Summary

Building Multi-Agent Environments

Adversarial and cooperative self-play

Training self-play environments

Adversarial self-play

Multi-brain play

Adding individuality with intrinsic rewards

Extrinsic rewards for individuality

Creating uniqueness with customized reward functions

Configuring the agents' personalities

Exercises

Summary

Section 3: Building Games

Debugging/Testing a Game with DRL

Introducing the game

Setting up ML-Agents

Introducing rewards to the game

Setting up TestingAcademy

Scripting the TestingAgent

Setting up the TestingAgent

Overriding the Unity input system

Building the TestingInput

Adding TestingInput to the scene

Overriding the game input

Configuring the required brains

Time for training

Testing through imitation

Configuring the agent to use IL

Analyzing the testing process

Sending custom analytics

Exercises

Summary

Obstacle Tower Challenge and Beyond

The Unity Obstacle Tower Challenge

Deep Learning for your game?

Building your game

More foundations of learning

Summary

Other Books You May Enjoy

Leave a review - let other readers know what you think