Well, the algorithm is death and reproduction, isn't it? If the output of your networks performs well enough when it confronts the real world, you get to repro

Well, the algorithm is death and reproduction, isn't it? If the output of your networks performs well enough when it confronts the real world, you get to reproduce before you die, and some of the essence of your network gets to live for another generation. If not, you die, and your network doesn't get to contribute to the next generation.