You're reading from Deep Reinforcement Learning with Python Master classic RL, deep RL, distributional RL, inverse RL, and more with OpenAI Gym and TensorFlow

Product type Paperback

Published in Sep 2020

Publisher Packt

ISBN-13 9781839210686

Length 760 pages

Edition 2nd Edition

Languages

Python

Tools

Deep Reinforcement Learning

Concepts

Deep Reinforcement Learning

Author (1):

Sudharsan Ravichandiran

View More author details

Table of Contents (22) Chapters

Preface

1. Fundamentals of Reinforcement Learning

2. A Guide to the Gym Toolkit FREE CHAPTER

3. The Bellman Equation and Dynamic Programming

4. Monte Carlo Methods

5. Understanding Temporal Difference Learning

6. Case Study – The MAB Problem

7. Deep Learning Foundations

8. A Primer on TensorFlow

9. Deep Q Network and Its Variants

10. Policy Gradient Method

11. Actor-Critic Methods – A2C and A3C

12. Learning DDPG, TD3, and SAC

13. TRPO, PPO, and ACKTR Methods

14. Distributional Reinforcement Learning

15. Imitation Learning and Inverse RL

16. Deep Reinforcement Learning with Stable Baselines

17. Reinforcement Learning Frontiers

18. Other Books You May Enjoy

19. Index

Appendix 1 – Reinforcement Learning Algorithms

1. Appendix 2 – Assessments

The Bellman equation

The Bellman equation, named after Richard Bellman, helps us solve the Markov decision process (MDP). When we say solve the MDP, we mean finding the optimal policy.

As stated in the introduction of the chapter, the Bellman equation is ubiquitous in reinforcement learning and is widely used for finding the optimal value and Q functions recursively. Computing the optimal value and Q functions is very important because once we have the optimal value or optimal Q function, then we can use them to derive the optimal policy.

In this section, we'll learn what exactly the Bellman equation is and how we can use it to find the optimal value and Q functions.

The Bellman equation of the value function

The Bellman equation states that the value of a state can be obtained as a sum of the immediate reward and the discounted value of the next state. Say we perform an action a in state s and move to the next state and obtain a reward r, then...