You're reading from Deep Reinforcement Learning Hands-On Apply modern RL methods to practical problems of chatbots, robotics, discrete optimization, web automation, and more

Product type Paperback

Published in Jan 2020

Publisher Packt

ISBN-13 9781838826994

Length 826 pages

Edition 2nd Edition

Languages

Python

Tools

Deep Reinforcement Learning

Concepts

Chatbots

Author (1):

Maxim Lapan

View More author details

Table of Contents (28) Chapters

Preface

1. What Is Reinforcement Learning?

2. OpenAI Gym FREE CHAPTER

3. Deep Learning with PyTorch

4. The Cross-Entropy Method

5. Tabular Learning and the Bellman Equation

6. Deep Q-Networks

7. Higher-Level RL Libraries

8. DQN Extensions

9. Ways to Speed up RL

10. Stocks Trading Using RL

11. Policy Gradients – an Alternative

12. The Actor-Critic Method

13. Asynchronous Advantage Actor-Critic

14. Training Chatbots with RL

15. The TextWorld Environment

16. Web Navigation

17. Continuous Action Space

18. RL in Robotics

19. Trust Regions – PPO, TRPO, ACKTR, and SAC

20. Black-Box Optimization in RL

21. Advanced Exploration

22. Beyond Model-Free – Imagination

23. AlphaGo Zero

24. RL in Discrete Optimization

25. Multi-agent RL

26. Other Books You May Enjoy

27. Index

The training process

Now that you know how the state of the cube is encoded in a 20 × 24 tensor, let's talk about the NN architecture and how it is trained.

The NN architecture

On the figure that follows (taken from the paper), the network architecture is shown.

Figure 24.2: The NN architecture transforming the observation (top) to the action and value (bottom)

As the input, it accepts the already familiar cube state representation as a 20 × 24 tensor and produces two outputs:

The policy, which is a vector of 12 numbers, representing the probability distribution over our actions.
The value, a single scalar estimating the "goodness" of the state passed. The concrete meaning of a value will be discussed later.

Between the input and output, the network has several fully connected layers with exponential linear unit (ELU) activations. In my implementation, the architecture is exactly the same as in the paper, and the model is...