This is an experimental web 3D exploring national selection. The problem domain is defined as a landscape with peaks and valleys. There are a number of solutions in the landscape which is defined as peaks. An initial population of creatures is randomly positioned in the landscape. In each generation, sexual reproduction occurs among a pair of creatures (cubes). Only the fittest reproduce. The weak (those with the highest cost - lower regions) are discarded and replaced by the new creatures reproduced from the parents. This Genetic Algorithm (GA) simulation demonstrates the survival of the fittest. In the simulation, the fittest creature turns red and emits flaming particles.
Click on the 'Start Sim' button in the viewport below to simulate the GA. Note that a higher population (>100) yields better solution.