Book Image

Hands-On Reinforcement Learning for Games

By : Micheal Lanham
Book Image

Hands-On Reinforcement Learning for Games

By: Micheal Lanham

Overview of this book

With the increased presence of AI in the gaming industry, developers are challenged to create highly responsive and adaptive games by integrating artificial intelligence into their projects. This book is your guide to learning how various reinforcement learning techniques and algorithms play an important role in game development with Python. Starting with the basics, this book will help you build a strong foundation in reinforcement learning for game development. Each chapter will assist you in implementing different reinforcement learning techniques, such as Markov decision processes (MDPs), Q-learning, actor-critic methods, SARSA, and deterministic policy gradient algorithms, to build logical self-learning agents. Learning these techniques will enhance your game development skills and add a variety of features to improve your game agent’s productivity. As you advance, you’ll understand how deep reinforcement learning (DRL) techniques can be used to devise strategies to help agents learn from their actions and build engaging games. By the end of this book, you’ll be ready to apply reinforcement learning techniques to build a variety of projects and contribute to open source applications.
Table of Contents (19 chapters)
1
Section 1: Exploring the Environment
7
Section 2: Exploiting the Knowledge
15
Section 3: Reward Yourself

Understanding imagination-augmented agents

The concept of imagination-augmented agents (I2A) was released in a paper titled Imagination-Augmented Agents for Deep Reinforcement Learning in February 2018 by T. Weber, et al. We have already talked about why imagination is important for learning and learning to learn. Imagination allows us to fill in the gaps in our learning and make leaps in our knowledge, if you will.

Giving agents an imagination allows us to combine model-based and model-free learning. Most of the agent algorithms we have used in this book have been model-free, meaning that we have no representative model of the environment. Early on, we did cover model-based RL with MC and DP, but most of our efforts have been fixed on model-free agents. The benefit of having a model of the environment is that the agent can then plan. Without a model, our agent just becomes reactionary...