Summary

In this chapter, we explored some practical examples of the application of technologies based on reinforcement learning.

First, we addressed the DeepMind AlphaZero project. AlphaZero is an artificial-intelligence algorithm based on machine learning techniques developed by Google DeepMind. It is a generalization of AlphaGo Zero, the predecessor developed specifically for the game of Go and in turn the evolution of AlphaGo, the first software capable of achieving superhuman performances in the game of Go.

Then, we explored the IBM Watson project. This is a system of artificial intelligence, able to answer questions expressed in natural language, developed within the DeepQA project of IBM by the research team directed by David Ferrucci. We looked at Unity's ML-Agents toolkit, FANUC industrial robots, and an automated trading system.

Finally, we discussed some future...