Abstract
During the past 35 years the evolutionary computation research community has been studying properties of evolutionary algorithms. Many claims have been made---these varied from a promise of developing an automatic programming methodology to solving virtually any optimization problem (as some evolutionary algorithms are problem independent). However, the most important claim was related to applicability of evolutionary algorithms to solving very complex business problems, i.e. problems, where other techniques failed. So it might be worthwhile to revisit this claim and to search for evolutionary algorithm-based software applications, which were accepted by businesses and industries. In this article Zbigniew Michalewicz attempts to identify reasons for the mismatch between the efforts of hundreds of researchers who make substantial contribution to the field of evolutionary computation and the number of real-world applications, which are based on concepts of evolutionary algorithms.
Index Terms
- Ubiquity symposium: Evolutionary computation and the processes of life: the emperor is naked: evolutionary algorithms for real-world applications
Recommendations
Ubiquity symposium: Evolutionary computation and the processes of life: the essence of evolutionary computation
In this third article in the ACM Ubiquity symposium on evolutionary computation Xin Yao provides a deeper understanding of evolutionary algorithms in the context of classical computational paradigms. This article discusses some of the most important ...
Ubiquity symposium: Evolutionary computation and the processes of life: life lessons taught by simulated evolution
In this second article in the ACM Ubiquity symposium on evolutionary computation Hans-Paul Schwefel explores the effects of simulating evolutionary mechanisms.
Ecological theory provides insights about evolutionary computation
GECCO '18: Proceedings of the Genetic and Evolutionary Computation Conference CompanionPromoting diversity in an evolving population is important for Evolutionary Computation (EC) because it reduces premature convergence on suboptimal fitness peaks while still encouraging both exploration and exploitation [3]. However, some types of ...
Comments