Evolutionary Computation

Computational techniques that mimic the process of biological evolution are considered to be “evolutionary computation”. This is an umbrella term that can include the areas known as; evolutionary/genetic algorithms, evolutionary/genetic programming, evolutionary/genetic strategies and more.

Biological evolution works through a somewhat random or haphazard insertion of mutations and other variations that produce diversification in biological organisms. The continually evolving set of organisms are constantly exposed to both survival threats and opportunities to merge and recombine into new threads. Weaker results get eliminated because they don’t survive to pass on their genetic material and stronger results do survive until eliminated or modified themselves.

Evolutionary or genetic algorithms are mathematical representations of the biological process, but not limited to problems involving biology. The algorithms are usually encoded into computer software programs that can be run continuously or by parallel systems in order to discover answers that would otherwise take a long time using the same methods without the speed of computers.

A simplified way to describe these algorithms is to say that they try all possible answers, eliminating the ones that produce false results until a correct answer is found, but they do this fast enough to make the approach practical to attempt. In actual practice, complex techniques are used to make decisions on how to pursue the branches of possibilities.

Leave a Reply

You must be logged in to post a comment.