A differential invasive weed optimization algorithm for improved global numerical optimization

Hybridization in context to Evolutionary Computation (EC) aims at combining the operators and methodologies from different EC paradigms to form a single algorithm that may enjoy a statistically superior performance on a wide variety of optimization problems. In this article we propose a simple but very efficient hybrid evolutionary algorithm that embeds the difference vector based mutation scheme of Differential Evolution (DE) into another recently developed global optimization algorithm known as Invasive Weed Optimization (IWO). IWO emulates the ecological behavior of the colonizing weeds. The hybrid algorithm, referred by us as Differential Invasive Weed Optimization (DIWO), is shown to possess greater explorative power as compared to the original DE and original IWO, through an analysis of the change of population-variance of these algorithms over successive generations and also through empirical simulations. We compare DIWO with original IWO, a modified IWO, two best known DE-variants: SaDE and JADE, and two state-of-the-art real optimizers: G-CMA-ES (Restart Covariance Matrix Adaptation Evolution Strategy with increasing population size) and DMS-PSO (Dynamic Multi Swarm Particle Swarm Optimization) over a test-suite of 25 shifted, rotated, and compositional benchmark functions and also one engineering optimization problem. Our comparative study indicates that although the hybridization scheme does not impose any serious burden on DIWO in terms of number of Function Evaluations (FEs), DIWO still enjoys a statistically superior performance over most of the tested benchmarks and especially over the multi-modal, rotated, and compositional ones in comparison to the other algorithms considered here.