The Reinforcement Learning Workshop

封面

版权信息

Preface

1. Introduction to Reinforcement Learning

Introduction

Learning Paradigms

Fundamentals of Reinforcement Learning

Reinforcement Learning Frameworks

Applications of Reinforcement Learning

Summary

2. Markov Decision Processes and Bellman Equations

Introduction

Markov Processes

3. Deep Learning in Practice with TensorFlow 2

Introduction

An Introduction to TensorFlow and Keras

How to Implement a Neural Network Using TensorFlow

Simple Regression Using TensorFlow

Simple Classification Using TensorFlow

TensorBoard – How to Visualize Data Using TensorBoard

Summary

4. Getting Started with OpenAI and TensorFlow for Reinforcement Learning

Introduction

OpenAI Gym

OpenAI Universe – Complex Environment

TensorFlow for Reinforcement Learning

OpenAI Baselines

Training an RL Agent to Solve a Classic Control Problem

Summary

5. Dynamic Programming

Introduction

Solving Dynamic Programming Problems

Identifying Dynamic Programming Problems

Dynamic Programming in RL

Summary

6. Monte Carlo Methods

Introduction

The Workings of Monte Carlo Methods

Understanding Monte Carlo with Blackjack

Types of Monte Carlo Methods

Exploration versus Exploitation Trade-Off

Importance Sampling

Solving Frozen Lake Using Monte Carlo

Summary

7. Temporal Difference Learning

Introduction to TD Learning

TD(0) – SARSA and Q-Learning

N-Step TD and TD(λ) Algorithms

The Relationship between DP Monte-Carlo and TD Learning

Summary

8. The Multi-Armed Bandit Problem

Introduction

Formulation of the MAB Problem

The Python Interface

The Greedy Algorithm

The Explore-then-Commit Algorithm

The ε-Greedy Algorithm

The UCB algorithm

Thompson Sampling

Contextual Bandits

Summary

9. What Is Deep Q-Learning?

Introduction

Basics of Deep Learning

Basics of PyTorch

The Action-Value Function (Q Value Function)

Deep Q Learning

Challenges in DQN

Summary

10. Playing an Atari Game with Deep Recurrent Q-Networks

Introduction

Understanding the Breakout Environment

CNNs in TensorFlow

Combining a DQN with a CNN

RNNs in TensorFlow

Building a DRQN

Introduction to the Attention Mechanism and DARQN

Summary

11. Policy-Based Methods for Reinforcement Learning

Introduction

Policy Gradients

Deep Deterministic Policy Gradients

Improving Policy Gradients

Summary

12. Evolutionary Strategies for RL

Introduction

Problems with Gradient-Based Methods

Introduction to Genetic Algorithms

Summary

Appendix

1. Introduction to Reinforcement Learning

2. Markov Decision Processes and Bellman Equations

3. Deep Learning in Practice with TensorFlow 2

4. Getting started with OpenAI and TensorFlow for Reinforcement Learning

5. Dynamic Programming

6. Monte Carlo Methods

7. Temporal Difference Learning

8. The Multi-Armed Bandit Problem

9. What Is Deep Q-Learning?

10. Playing an Atari Game with Deep Recurrent Q-Networks