Quantum strategy of population initialization in genetic algorithm

Quantum Genetic Algorithm is a relatively new field of study to enhance the computational efficiency of the Darwinian optimization process in genetic algorithms with quantum speedup techniques. This paper introduces an application strategy of the quantum counting algorithm to genetic algorithms, particularly aimed to enhance the initial population setup at the beginning of optimization. More specifically, our goal is to exploit a quantum algorithm to count the number of marked items from an unstructured list quadratically faster than classical algorithms in order to detect the presence and amount of unsuitable individuals in a stochastically generated initial population, thereby starting optimization with a mark of potential to improve the performance in the later stage. The advantage of our method is examined via a conventional genetic algorithm to solve the 0-1 Knapsack problem with varying cases of the constraints, and a comparative analysis on the optimizing performance is made accordingly.