Chapter 17
Playing with Deep Reinforcement Learning
IN THIS CHAPTER
Presenting reinforcement learning
Using OpenAI Gym for experimentation
Determining how a Deep Q-Network (DQN) works
Working with AlphaGo, AlphaGo Zero, and Alpha Zero
Apart from the example of GANs, you may be tempted to identify deep learning with supervised learning predictions. However, you also use deep learning for unsupervised learning and reinforcement learning (RL). Unsupervised learning supports a number of established techniques, such as autoencoders and self-organizing maps (SOMs), which this book doesn’t cover. Unsupervised techniques can help you to segment your data into homogeneous groups or to detect anomalies in your variables.
RL techniques are even more popular than unsupervised learning techniques among practitioners. Recently the object of intense research, RL achieves smarter solutions for problems such as parking a car, learning to drive in as little as twenty minutes (as this paper illustrates: https://arxiv.org/abs/1807.00412), controlling an industrial robot, and more. (This article by ...