Search icon CANCEL
Subscription
0
Cart icon
Your Cart (0 item)
Close icon
You have no products in your basket yet
Save more on your purchases now! discount-offer-chevron-icon
Savings automatically calculated. No voucher code required.
Arrow left icon
Explore Products
Best Sellers
New Releases
Books
Videos
Audiobooks
Learning Hub
Conferences
Free Learning
Arrow right icon
Arrow up icon
GO TO TOP
Hands-On Reinforcement Learning for Games

You're reading from   Hands-On Reinforcement Learning for Games Implementing self-learning agents in games using artificial intelligence techniques

Arrow left icon
Product type Paperback
Published in Jan 2020
Publisher Packt
ISBN-13 9781839214936
Length 432 pages
Edition 1st Edition
Languages
Arrow right icon
Author (1):
Arrow left icon
Micheal Lanham Micheal Lanham
Author Profile Icon Micheal Lanham
Micheal Lanham
Arrow right icon
View More author details
Toc

Table of Contents (19) Chapters Close

Preface 1. Section 1: Exploring the Environment
2. Understanding Rewards-Based Learning FREE CHAPTER 3. Dynamic Programming and the Bellman Equation 4. Monte Carlo Methods 5. Temporal Difference Learning 6. Exploring SARSA 7. Section 2: Exploiting the Knowledge
8. Going Deep with DQN 9. Going Deeper with DDQN 10. Policy Gradient Methods 11. Optimizing for Continuous Control 12. All about Rainbow DQN 13. Exploiting ML-Agents 14. DRL Frameworks 15. Section 3: Reward Yourself
16. 3D Worlds 17. From DRL to AGI 18. Other Books You May Enjoy

Understanding policy gradient methods

One thing we need to understand about PG methods is why we need them and what the intuition is behind them. Then, we can cover some of the mathematics very briefly before diving into the code. So, let's cover the motivation behind using PG methods and what they hope to achieve beyond the other previous methods we have looked at. I have summarized the main points of why/what PG methods do and try to solve:

  • Deterministic versus stochastic functions: We often learn early in science and mathematics that many problems require a single or deterministic answer. In the real world, however, we often equate some amount of error to deterministic calculations to quantify their accuracy. This quantification of how accurate a value is can be taken a step further with stochastic or probabilistic methods.

Stochastic methods are often used to quantify...

lock icon The rest of the chapter is locked
Register for a free Packt account to unlock a world of extra content!
A free Packt account unlocks extra newsletters, articles, discounted offers, and much more. Start advancing your knowledge today.
Unlock this book and the full library FREE for 7 days
Get unlimited access to 7000+ expert-authored eBooks and videos courses covering every tech area you can think of
Renews at $19.99/month. Cancel anytime