论文部分内容阅读
针对传统贝叶斯优化算法进化效率低及收敛速度慢的情况,提出一种新型混合贝叶斯优化算法.该算法利用适应度遗传及个体的局部搜索方法,使种群个体趋向于全局最优解,提高了进化效率.为提高贝叶斯优化算法中贝叶斯网络结构学习的效率,提出一种爬山法和模式蚁群算法相结合的网络结构学习方法,同时对新型贝叶斯优化算法的收敛性进行了分析.利用典型的函数对提出的新型混合贝叶斯优化算法进行了仿真分析,证明了所提出的方法可以有效地加快算法的收敛速度和收敛精度.将该算法应用于目标分配问题中,仿真证明了所提算法的有效性和优越性.
Aiming at the low evolutionary efficiency and slow convergence speed of traditional Bayesian optimization algorithm, a new hybrid Bayesian optimization algorithm is proposed, which uses the genetic algorithm of fitness and the local search method of individual to make the population tend to the global optimal solution , Which improves the efficiency of evolutionary.In order to improve the efficiency of Bayesian network structure learning in Bayesian optimization algorithm, a new method of network structure learning based on hill-climbing method and model ant colony algorithm is proposed. Meanwhile, the new Bayesian optimization algorithm The convergence of the new hybrid Bayesian optimization algorithm is simulated and analyzed by using the typical function.The simulation results show that the proposed method can effectively speed up the convergence speed and the convergence precision of the algorithm.The algorithm is applied to the target allocation In the problem, the simulation proves the validity and superiority of the proposed algorithm.