IVF/NSGAII: In Vitro Fertilization Method Coupled to NSGAII

The In Vitro Fertilization Genetic Algorithm - IVF/GA is a promising algorithm applicable to mono-objective problems, especially for complex and multimodal ones. Due to the balance between the exploration and exploitation of the IVF/GA, and thus its abilities to avoid the local optimum, we speculate that the IVF can also improve the Multi-Objective Evolutionary Algorithms (MOEA). So, this paper proposes the IVF/NSGAII, which is the IVF coupled to NSGAII. We evaluate the proposal's improvement comparing the canonical NSGAII versus IVF/NSGAII, applied to Zitzler-Deb-Thiele (ZDT) functions. The results show the IVF/NSGAII outperformed the canonical version on Inverted Generational Distance (IGD) metric, mainly in the complex functions, e.g. ZDT 4, 5 and 6. Therefore, we conclude that IVF improved the NSGAII and also is a promising approach to support MOEA's.