Our journey through this book has been an exploration of the evolution of reinforcement and deep reinforcement learning (DRL). We have looked at many methods that you can use to solve a variety of problems in a variety of environments, but in general, we have stuck to a single environment; however, the true goal of DRL is to be able to build an agent that can learn across many different environments, an agent that can generalize its knowledge across tasks, much like we animals do. That type of agent, the type that can generalize across multiple tasks without human intervention, is known as an artificial general intelligence, or AGI. This field is currently exploding in growth for a variety of reasons and will be our focus in this final chapter.
In this chapter, we will look at how DRL builds the AGI agent. We will first look at the concept of meta learning, or...
