I. Introduction
Optimization is the process of finding a set of solutions in an unknown solution space so that the objective to be solved achieves optimal performance [1] . It is widespread in engineering, physics, economics, biology, etc. Several traditional optimization algorithms have shown outperformance on some classical optimization problems. Monte Carlo simulations can find a near-native protein conformation in small experimental protein landscape [2] . Lagrange multiplier optimization improves the operating efficiency of energy equipment and its high-performance computing satisfies the real-time requirements of actual engineering [3] . K-means, a classical clustering algorithm, is essentially an optimization algorithm that regards the samples as the solution space and takes accurate classification as the optimal objective. It demonstrates as remarkable performance in some simple medical sample classification tasks [4] .