Summary
In this chapter, we discussed discrete optimization problems – a subfield of the optimization domain that deals with discrete structures like graphs or sets. We checked RL’s applicability using the Rubik’s cube as a well-known, but still challenging, problem. But in general, this topic is much wider than just puzzles – the same methods could be used in optimizing schedules, optimal route planning, and other practical topics.
In the final chapter of the book, we will talk about multi-agent problems in RL.
