Genetic Algorithms: Theory And Application
Edgar N. Reyes, Dennis I. Merino, Carl W. Steidley · 2020
Abstract NOTE: The first page of text has been automatically extracted and included below in lieu of an abstract Session 2220 Genetic Algorithms: Theory and Application Dennis I. Merino, Edgar N. Reyes/Carl Steidley Southeastern Louisiana University/Texas A&M University - Corpus Christi Hammond, LA 70402/ Corpus Christi, TX 78412 1 Introduction Genetic algorithms, a class of robust and efficient search techniques that can be randomly sample large spaces, have applications in the field of optimization and in a wide range of computer science problems in pattern recognition, search, scheduling, and machine learning. Genetic algorithms are motivated by characteristics found in natural population genetics, among them robustness and efficiency. Features of biological systems found in genetic algorithms include reproduction, self- guidance, self-repair, the nature of survival of the fittest, and variation through mutation. Genetic algorithms were developed by John Holland of the University of Michigan in the 1970's. Many of the essential properties of genetic algorithms discussed in this paper can be found in [1, 2]. When a genetic algorithm is used to find an optimal solution in the space of all feasible solutions, the algorithm maintains a population (or set) of feasible solutions which evolve through random process based on principles found in the mechanics of natural selection and genetics. Each time this set of solution evolves (or as we say iterated), a new set of solutions is generated. The iterations are repeated several times as necessary. An important goal from the standpoint of genetic algorithms is to improve the values of the objective function in an efficient, practical, or quick method. Ideally, as it is true for many optimization techniques, one would like to generate after several iterations a set of solutions that includes an optimal global solution; but a genetic algorithm has its priority set on tying improvement of an objective function with performance-practicality of the method. In the literature, one finds several applications of genetic algorithms in routing and scheduling problems, machine learning, setting weights in neural nets, and testing expert systems that control complex systems, amongst others [2]. In the next section, we focus on the algorithm itself by solving an elementary optimization problem. 2 Genetic Algorithm: An example We present the following to illustrate the effectiveness of genetic algorithms. We note that although the problem can be solved easily using a different method, it is the process involved with genetic algorithms that is put to focus. We wish to