元启发式优化算法是一种解决全局优化问题常用的方法,它主要是通过模拟自然和人类智慧来实现最优解的求解。
相比于传统的优化方法,如模拟退回,梯度下降等,1960年。元启发式优化方法首次被提出,是一种灵活且无视梯度变化的方法。
元启发式的优化算法主要可被分为四类:
(1)基于进化的算法
(2)基于群体智能的算法
(3)基于人类的算法
(4)基于物理和化学的算法
下面我们逐步介绍一下上述四种主要的智能优化算法。
基于进化的算法主要是通过模拟自然界中的优胜劣汰的进化法则(达尔文法则),实现种群的整体进步,最终完成最优解的求解。其中以遗传算法(Genetic Algorithm,GA)和差分进化(Differential Evolution ,DE)为主要代表。
随着科学家对基于自然进化算法的不断探索,随后也提出了多种进化优化算法,如:
evolutionary strategy (ES),https://www.scirp.org/reference/referencespapers.aspx?referenceid=1536208
evolutionary programming(EP),https://ieeexplore.ieee.org/document/771163/metrics#metrics
gene expression programming (GEP),https://www.researchgate.net/publication/285414114_Gene_Expression_Programming-A_New_Adaptive_Algorithm_for_Solving_Problems
genetic programming(GP),附上一篇相似文章吧,原文没找到,有的网友告知啊,https://link.springer.com/chapter/10.1007/0-387-28356-0_5
covariance matrix adaptation evolution strategy(CMA-ES),https://link.springer.com/chapter/10.1007/978-3-540-30217-9_29
biogeographybased optimization(BBO),https://ieeexplore.ieee.org/abstract/document/4475427/
Genetic Algorithm(GA),https://deepblue.lib.umich.edu/bitstream/handle/2027.42/46947/10994_2005_Article_422926.pdf
Differential Evolution (DE),https://link.springer.com/article/10.1023/A:1008202821328
群体智能优化算法是通过模拟群体的智慧,来实现全局最优解的获取。在该算法中,每一个群体都是一个生物种群,通过种群中个体之间的协同行为,进而完成个体无法完成的任务。下面将列出一部分基于群体智能的优化算法。
如:
particle swarm optimization (PSO),https://www.researchgate.net/publication/24293528_Particle_Swarm_Optimization
ant colony optimization (ACO) ,https://ieeexplore.ieee.org/document/6281178?reload=true&tp=&arnumber=6281178
artificial bee colony (ABC),
https://www.researchgate.net/publication/281168311_An_artificial_bee_colony_ABC_algorithm_for_numeric_function_optimization_In_Proceedings_of_the_IEEE_swarm_intelligence_symposium_Indianapolis_IN_USA
bacterial foraging(BF),https://ieeexplore.ieee.org/document/1004010
bat algorithm (BA), https://www.scienceopen.com/document?vid=851e27f2-134f-4d9b-b08e-38bfb7af65ce
firefly algorithm (FFA),https://www.oalib.com/paper/3945747#.YDS12sD3Fbw
krill herb (KB),https://www.sciencedirect.com/science/article/abs/pii/S1007570412002171
cuckoo search (CS),https://ieeexplore.ieee.org/document/5393690
monkey search (MS),https://aip.scitation.org/doi/10.1063/1.2817338
bee colony optimization (BCO) ,https://www.researchgate.net/publication/284410603_Bee_colony_optimization_-_A_cooperative_learning_approach_to_complex_transportation_problems
cat swarm,https://ieeexplore.ieee.org/document/4620980/
wolf search (WS),https://ieeexplore.ieee.org/document/6360147
ant lion optimizer (ALO),https://www.sciencedirect.com/science/article/abs/pii/S0965997815000113
grey wolf optimization (GWO),https://www.sciencedirect.com/science/article/abs/pii/S0965997813001853
whale-optimization algorithm (WOA),https://www.sciencedirect.com/science/article/abs/pii/S0965997816300163
crow search algorithm (CSA),https://www.sciencedirect.com/science/article/abs/pii/S0045794916300475
Salp swarm algorithm (SSA),https://www.sciencedirect.com/science/article/abs/pii/S0957417421001263
grasshopper optimization algorithm (GOA) ,https://www.sciencedirect.com/science/article/abs/pii/S0965997816305646
butterfly optimization algorithm (BOA) ,https://link.springer.com/article/10.1007/s00500-018-3102-4
squirrel search algorithm (SSA),https://www.sciencedirect.com/science/article/abs/pii/S2210650217305229
Harris Hawks optimization (HHO),https://www.sciencedirect.com/science/article/abs/pii/S0167739X18313530
这一届主要介绍的是基于进化的算法和基于群体智能的算法,并且给出了所出的网页链接,可供学友们查看和下载。
下一节中将介绍基于人类的算法和基于物理和化学的算法的元启发式优化算法。