In this chapter, you will learn about an advanced solution search optimization method that can be used to create autonomous navigator agents. This method is called Novelty Search (NS). The main idea of this method is that an objective function can be defined using the novelty of the behavior exposed by the solver agent, rather than the distance to a goal in the solution search space.
In this chapter, you will learn how to use NS-based search optimization methods with the neuroevolution algorithm to train successful maze navigation agents. By conducting the experiments presented in this chapter, you will also see that the NS method is superior to the conventional goal-oriented search optimization method for specific tasks. By the end of this chapter, you will have learned the basics of the NS optimization method. You will be able to define the...
