A computationally efficient evolutionary algorithm for real-parameter optimization
Evolutionary Computation
Differential Evolution: A Practical Approach to Global Optimization (Natural Computing Series)
Differential Evolution: A Practical Approach to Global Optimization (Natural Computing Series)
Development of efficient particle swarm optimizers by using concepts from evolutionary algorithms
Proceedings of the 12th annual conference on Genetic and evolutionary computation
Hi-index | 0.00 |
In past, only a few attempts have been made in adopting a unified outlook towards different paradigms in Evolutionary Computation. The underlying motivation of these studies was aimed at gaining better understanding of evolutionary methods, both at the level of theory as well as application, in order to design efficient evolutionary algorithms for solving wide-range complex problems. One such attempt is made in this paper, where we reinstate 'Unified Theory Of Evolutionary Computation', drawn from past studies, and investigate four steps - Initialization, Selection, Generation and Replacement, which are sufficient to describe common Evolutionary Optimization Systems such as Genetic Algorithms, Evolutionary Strategies, Evolutionary Programming, Particle Swarm Optimization and Differential Evolution. As a next step we consider Differential Evolution, a relatively new evolutionary paradigm, and discover its inability to efficiently solve unimodal problems when compared against a benchmark Genetic Algorithm. Targeted towards enhancing DE's performance, several modifications are successfully proposed and validated through simulation results. The Unified Approach is found helpful in understanding the role and re-modeling of DE steps to efficiently solve unimodal problems.