Genetic Algorithms: A Technical Implementation of Natural Evolution

It is now inevitable to accept the fact that evolution is the reason why species now have not only sustained and adapted to the environment but also have developed the intelligence to improve their lifestyle to make it more interesting, livable, and safer for mostly all kinds of threats. Another feature of evolution is that it never goes backward as in every case, the fittest individual is the chosen one for survival. Although this theory is confidently backed by many discoveries in science, many conspiracy theories in contrast to evolution still prevail. Whether or not evolution transpired is still a matter of debate, but one cannot neglect the fact that this theory can be deployed in the field of artificial intelligence to develop algorithms and smart enough to yield the desired results. The algorithms are now being implemented in both machine learning and deep learning environments. A significant amount of the research done here belongs to the applications of unique genetic algorithms depending on the scenario. These algorithms are called evolutionary algorithms of which a major slice belongs to the genetic algorithms. This is an algorithm which inputs a set of initial individuals and performs three operators to produce better and smarter individuals after every generation. The three operators are selection, crossover, and mutation. This chapter is a holistic review of the genetics algorithm mechanism.