Evolutionary Computation: The Profitable Gene

Abstract

Another key source of enhancing human intelligence is inspired by evolution in nature. Biological evolution has been particularly successful in the design and creation of amazingly complex organisms driven by several simple mechanisms. According to Darwin, the driving force behind natural evolution is the capability of a population of individuals to reproduce and deliver new populations of individuals, which are fitter for their environment. The fundamental evolutionary step is survival of the fittest, which implies some sort of competition, combined with recombination acting on the chromosomes, rather than on the living organisms themselves. Evolutionary computation uses an analogy with natural evolution to perform a search by evolving solutions (equations, electronic schemes, mechanical part, etc.) to problems in the virtual environment of computers. One of the important features of evolutionary computation is that instead of working with one solution at a time in the search-space, a large collection or population of solutions is considered at once. The better solutions are allowed to “have children” and the worse solutions are quickly eliminated. The “child solutions” inherit their “parents' characteristics” with some random variation, and then the better of these solutions are allowed to “have children” themselves, while the worse ones “die”, and so on. This simple procedure causes simulated evolution. After a number of generations the computer will contain solutions which are substantially better than their long-dead ancestors at the start.