Hands-On Neuroevolution with Python

Hyperparameter selection

The hyperparameters are defined as a Parameters Python class, and the MultiNEAT library refers to it for the necessary configuration options. In the source code of the experiment runner script, we define a specialized function called create_hyperparameters, which encapsulates the logic of the hyperparameter initialization. Hereafter, we describe the most critical hyperparameters and the reasons for choosing these specific values:

1. We decided to use a medium size for the CPPN genome population. This is done to intensify the evolution by providing a large space of options for the solution search from the beginning. The size of the population is defined as follows:

    params.PopulationSize = 300

2. Next, we define the number ...