Quantum gate optimization in a meta-level genetic quantum algorithm

Genetic quantum algorithms (GQA) are population based evolutionary algorithms that imitate quantum physics by introducing quantum bits for a basic probabilistic genotypic representation and hence better population diversity, and quantum gates for evolving the population of solutions. While quantum inspired gates play an important role in the evolutionary process, there is no specific method for their design, i.e. they are mostly developed through ad hoc procedures. Here, we propose a multi-objective meta-level GQA in order to determine parameters of QG which is applicable for a wide variety of optimization problems. Specifically, a two-layer GQA is constructed, in which the lower layer´s objective is to optimize the four junction types: Dejong, Peak, Easoms, and Griewank. And the higher layer´s objective is to determine optimal parameters for QG that helps the proposed algorithm find best solutions. GQA optimization performance, with optimized parameter of QG is compared with GA on several benchmark problems and the superiority of the proposed method is statistically shown.