A Study on Different Learning Strategies on Evolutionary Computation Techniques

Evolutionary computation (EC) techniques are a category of meta-heuristic learning algorithm. These are frequently employed to address issues with global optimization across numerous fields. The inherent natural search capabilities and interactions among the variables of such kind of algorithms make them optimized the objective functions of the investigated problem. Each evolutionary computation algorithm has its own learning strategy. In this work, a comprehensive study is made to investigate, compare, and understand different learning strategies in few classical and widely used ECs such as PSO, DE, TLBO, and SGO. This study will provide useful guidelines for further exploration and possible modifications of learning strategies for future applications. In addition, in this work, we have compared the challenges and applications of the above EC algorithms.