Learning-based monarch butterfly optimization algorithm for solving numerical optimization problems

Many optimization algorithms have been used to solve complex real-world problems, motivating many scientists to improve various optimization algorithms. The monarch butterfly optimization algorithm MBO has been proven to have good performance and is an effective tool for solving many problems. However, the migration and updating operators mainly learn from the global best and are often trapped in local optima as well as premature convergence in many optimization problems. This work introduces an adaptive learning strategy and elitism strategy namely ALMBO to enhance the exploration capacity of the algorithm and ensure that the fittest individuals are retained in the next generation. The experimental results on fourteen benchmark functions show that compared with the two representative MBO algorithm and other methods, ALMBO performs much better than the others on mean values, best values, standard deviation and convergence speed.