X. Ma et al., "A Multiobjective Evolutionary Algorithm Based on Decision Variable Analyses for Multiobjective Optimization Problems With Large-Scale Variables," in IEEE Transactions on Evolutionary Computation, vol. 20, no. 2, pp. 275-298, April 2016, doi: 10.1109/TEVC.2015.2455812.
的论文学习笔记,只供学习使用,不作商业用途,侵权删除。并且本人学术功底有限如果有思路不正确的地方欢迎批评指正!定义1 表示可以通过逐一优化变量来解决可分离函数。 可分离性意味着每个变量都独立于任何其他变量。 关于可分离函数和不可分离函数的其他定义可以在[7]和[12]中找到。 球面函数sphere function,广义Rastrigin函数,广义Griewank函数和Ackley函数[15],[16]是可分离函数的代表。 基本上,可分离性函数意味着可以独立于任何其他变量来优化问题中涉及的决策变量,而不可分离性函数意味着至少两个决策变量之间存在相互作用interactions。
变量依赖性是问题的重要方面,它们描述了问题的结构。如果预先知道问题的变量依赖性,则很容易将决策变量划分为几个子组件。因此,通过分别优化具有低维子组件的几个较简单的子问题来解决高维变量的难题是十分有益的。但是,问题的变量依赖性通常是事先未知的。此外,“相互依存变量”的定义不是唯一的。 Yu等[17]提出,当且仅当没有两个决策变量所携带的信息,关联子问题不能被优化时,两个决策变量相互作用。 Weise等[5]提出,如果改变一个决策变量对适应度的影响依赖于另一个决策变量的值,则两个决策变量会相互影响。与以上两个定性定义不同,本文使用以下相互依赖变量的定量定义。
定义2可以从Yang等人[7]提出的不可分函数的定义中得出。在这些“相互依赖变量”的不同定义中,我们选择“定义2”作为相互依赖变量的定义 因为该定义是 定量的 且易于使用。
m-1
,而距离变量的总数是n-m+1
定义二
来学习两个决策变量之间的交互关系,算法2给出了实现细节,图5-7展示了ZDT1,DTLZ1,UF1 和UF8以及五个WFG问题的两个决策变量之间的交互关系进化种群的结构如图11所示,其中N是种群大小。在本文中,MOEA / DVA优化了单个进化种群,所有子组件共享相同的种群。种群中的每个体都代表一个MOP。在此图中,我们假设x1和x2是多样化变量(位置变量或混合变量),而x3,x4,…。 。 。 ,x8是距离变量。距离变量分为三个独立的子分量{x3},{x4,x5,x6}和{x7,x8}。在算法优化早期,固定位置变量,只优化距离变量。sub-MOP的特征之一是多样性变量的值在演化的早期是固定的。因此,种群中多样性变量的分布对获得的解的分布具有重要影响。为了保证进化种群的多样性,均匀采样方法[37]被用来初始化种群多样性变量的值。
表III列出了多样性变量的分解与距离变量的分解之间的差异。 基于学习到的变量链接,MOEA / DVA通过算法3将距离变量分解为一组低维子组件。算法4提供了MOEA / DVA的详细信息。 在MOEA / DVA中,这两个分解共同解决了MOP。 MOEA / DVA首先通过具有均匀分布的多样性变量将困难的MOP分解为多个更简单的子MOP。 然后,每个子MOP在发展的早期阶段都逐一优化子组件。
算法4的第5行对距离变量进行分组。 算法4的第9-12行类似于协作协同进化框架[7],[21]。 算法4的第11行执行下一段介绍的子组件优化。 为简单起见,我们为MOEA / DVA中的每个子组件分配相同的计算资源。 也可以根据不同的子组件的近期性能为它们分配不同的计算资源[38]。
对于子组件优化,我们在MOEA/D [2]中使用进化算子。由于每个目标函数fi(x,i = 1,…,m)是连续的,因此相邻子MOP的最优解应该彼此接近。因此,有关其相邻子MOP的任何信息将有助于优化当前子MOP [2]。这些子MOP之间的邻域关系是基于其各个变量之间的欧几里得距离定义的。如果第i个子MOP的多样性变量接近第j个子MOP的多样性变量,则第i个子MOP是第j个子MOP的邻居。算法5提供了子组件优化的详细信息。在第3步中,由于在发展的早期阶段每个子MOP的多样性变量值是固定的,因此MOEA / DVA仅使用第i个MOP的后代来更新第i个MOP的当前解。为简单起见,算法5在发展的早期阶段为MOEA / DVA中的每个单独/子MOP分配了相同的计算资源。结论中将讨论更智能的MOEA / DVA版本。
最后,我们在算法4的第17行中介绍了在MOEA / DVA中进行均匀性优化的必要性。如上所述,MOEA / DVA首先将MOP分解为具有均匀分布的多样性变量的子MOP集合,并且逐个优化每个子MOP。 通过在演化的早期为多样性变量赋予统一固定值,MOEA / DVA可以在决策变量上通过均匀的多样性变量保持种群的多样性。 因此,MOEA / DVA找到的解决方案的分布高度依赖于问题从PS到PF的映射。
通过算法的效用评价算法的阶段–先固定多样性只对收敛性变量进行优化,然后当达到效用阈值,对所有变量进行优化
为了解决这一问题,利用MOEA/D对所有的决策变量进行演化,包括演化后期的多样性变量( if 效用 < 阈值 )。其目的是为了提高种群在目标空间中的均匀性。因此,MOEA/DVA的思想是逐个优化子组件,使进化种群在进化的早期阶段( if 效用 ≥ 阈值 )具有良好的收敛性。在进化后期,MOEA/DVA对包括不同变量在内的所有决策变量进行优化,使进化种群在目标空间中具有良好的均匀性。子组件优化的效用在算法6中计算。
[1] K. Deb, Multi-Objective Optimization Using Evolutionary Algorithms. New York, NY, USA: Wiley, 2001.
[2] Q. Zhang and H. Li, “MOEA/D: A multi-objective evolutionary algorithm based on decomposition,” IEEE Trans. Evol. Comput., vol. 11, no. 6, pp. 712–731, Dec. 2007.
[3] N. Beume, B. Naujoks, and M. Emmerich, “SMS-EMOA: Multiobjective selection based on dominated hypervolume,” Eur. J. Oper. Res., vol. 181, no. 3, pp. 1653–1669, 2007.
[4] K. Deb and H. Jain, “An evolutionary many-objective optimization algorithm using reference-point based non-dominated sorting approach, part I: Solving problems with box constraints,” IEEE Trans. Evol. Comput., vol. 18, no. 4, pp. 577–601, Aug. 2014.
[5] T. Weise, R. Chiong, and K. Tang, “Evolutionary optimization: Pitfalls and booby traps,” J. Comput. Sci. Technol., vol. 27, no. 5, pp. 907–936, 2012.
[6] M. Potter and K. Jong, “A cooperative coevolutionary approach to function optimization,” in Proc. Int. Conf. Parallel Probl. Solv. Nat., vol. 2. Jerusalem, Israel, 1994, pp. 249–257.
[7] Z. Yang, K. Tang, and X. Yao, “Large scale evolutionary optimization using cooperative coevolution,” Inf. Sci., vol. 178, no. 15, pp. 2985–2999, 2008.
[8] X. Li and X. Yao, “Cooperatively coevolving particle swarms for large scale optimization,” IEEE Trans. Evol. Comput., vol. 16, no. 2, pp. 210–224, Apr. 2012.
[9] Y. Mei, X. Li, and X. Yao, “Cooperative co-evolution with route distance grouping for large-scale capacitated arc routing problems,” IEEE Trans. Evol. Comput., vol. 18, no. 3, pp. 435–449, Jun. 2014.
[10] D. Goldberg, Genetic Algorithms in Search, Optimization, and Machine Learning. Reading, MA, USA: Addison-Wesley, 1989.
[11] Y. Chen, T. Yu, K. Sastry, and D. Goldberg, “A survey of linkage learning techniques in genetic and evolutionary algorithms,” Illinois Genet. Algorithms Libr., Univ. Illinois Urbana-Champaign, Urbana, IL, USA, Tech. Rep. 2007014, 2007.
[12] S. Huband, P. Hingston, L. Barone, and L. While, “A review of multiobjective test problems and a scalable test problem toolkit,” IEEE Trans. Evol. Comput., vol. 10, no. 5, pp. 477–506, Oct. 2006.
[13] Q. Zhang et al., “Multiobjective optimization test instances for the CEC 2009 special session and competition,” School Comput. Sci. Electr. Eng., Univ. Essex, Colchester, U.K., Tech. Rep. CES-887, 2008.
[14] Q. Zhang, A. Zhou, and Y. Jin, “RM-MEDA: A regularity model-based multiobjective estimation of distribution algorithm,” IEEE Trans. Evol. Comput., vol. 12, no. 1, pp. 41–63, Feb. 2008.
[15] L. Jiao, Y. Li, M. Gong, and X. Zhang, “Quantum-inspired immune clonal algorithm for global optimization,” IEEE Trans. Syst., Man, Cybern. B, Cybern., vol. 38, no. 5, pp. 1234–1253, Oct. 2008.
[16] K. Tang, X. Li, P. Suganthan, Z. Yang, and T. Weise, “Benchmark functions for the CEC’2010 special session and competition on large-scale global optimization,” Nat. Inspir. Comput. Appl. Lab., Univ. Sci. Technol. China, Hefei, China, Tech. Rep. 2010001, 2010.
[17] T. Yu, D. Goldberg, K. Sastry, C. Lima, and M. Pelikan, “Dependency structure matrix, genetic algorithms, and effective recombination,” Evol. Comput., vol. 17, no. 4, pp. 595–626, 2009.
[18] W. Chen, T. Weise, Z. Yang, and K. Tang, “Large-scale global optimization using cooperative coevolution with variable interaction learning,” in Proc. Conf. Parallel Probl. Solv. Nat., Kraków, Poland, 2010, pp. 300–309.
[19] P. Toint, “Test problems for partially separable optimization and results for the routine PSPMIN,” Dept. Math., Univ. Namur, Namur, Belgium, Tech. Rep. 83/4, 1983.
[20] B. Colson and P. Toint, “Optimizing partially separable functions without derivatives,” Optim. Methods Softw., vol. 20, nos. 4–5, pp. 493–508, 2005.
[21] M. Omidvar, X. Li, Y. Mei, and X. Yao, “Cooperative co-evolution with differential grouping for large scale optimization,” IEEE Trans. Evol. Comput., vol. 18, no. 3, pp. 378–393, Jun. 2014.
[22] D. Thierens and D. Goldberg, “Mixing in genetic algorithms,” in Proc. 5th Int. Conf. Genet. Algorithms, Urbana, IL, USA, 1993, pp. 38–45.
[23] M. Munetomo and D. Goldberg, “Identifying linkage groups by nonlinearity/ nonmonotonicity detection,” in Proc. Genet. Evol. Comput. Conf., vol. 1. Orlando, FL, USA, 1999, pp. 433–440.
[24] M. Tezuka, M. Munetomo, and K. Akama, “Linkage identification by nonlinearity check for real-coded genetic algorithms,” in Proc. Genet. Evol. Comput. Conf., Seattle, WA, USA, 2004, pp. 222–233.
[25] K. Weicker and N. Weicker, “On the improvement of coevolutionary optimizers by learning variable interdependencies,” in Proc. IEEE Congr. Evol. Comput., Washington, DC, USA, 1999, pp. 1627–1632.
[26] Y. Chen, “Extending the scalability of linkage learning genetic algorithms: Theory and practice,” Ph.D. dissertation, Dept. Comput. Sci., Univ. Illinois Urbana-Champaign, Urbana, IL, USA, 2004.
[27] J. E. Smith, “Self adaptation in evolutionary algorithms,” Ph.D. dissertation, Dept. Comput. Sci., Univ. West England, Bristol, U.K., 1998.
[28] Q. Zhang and H. Muehlenbein, “On the convergence of a class of estimation of distribution algorithms,” IEEE Trans. Evol. Comput., vol. 8, no. 2, pp. 127–136, Apr. 2004.
[29] G. R. Harik, F. G. Lobo, and D. E. Goldberg, “The compact genetic algorithm,” IEEE Trans. Evol. Comput., vol. 3, no. 4, pp. 287–297, Nov. 1999.
[30] M. Pelikan and D. Goldberg, “BOA: The Bayesian optimization algorithm,” in Proc. Genet. Evol. Comput. Conf., Orlando, FL, USA, 1999, pp. 525–532.
[31] T. Yu, A. Yassine, and D. Goldberg, “A genetic algorithm for developing modular product architectures,” in Proc. ASME Int. Design Eng. Tech. Conf., Chicago, IL, USA, 2003, pp. 515–524.
[32] K. Miettinen, Nonlinear Multiobjective Optimization. Boston, MA, USA: Kluwer Academic, 1999.
[33] C. Hillermeier, Nonlinear Multiobjective Optimization—A Generalized Homotopy Approach. Boston, MA, USA: Birkhauser, 2001.
[34] E. Zitzler, K. Deb, and L. Thiele, “Comparison of multiobjective evolutionary algorithms: Empirical results,” Evol. Comput., vol. 8, no. 2, pp. 173–195, 2000.
[35] K. Deb, L. Thiele, M. Laumanns, and E. Zitzler, “Scalable multiobjective optimization test problems,” in Proc. Congr. Evol. Comput., Honolulu, HI, USA, 2002, pp. 825–830.
[36] H. Liu, F. Gu, and Q. Zhang, “Decomposition of a multiobjective optimization problem into a number of simple multiobjective subproblems,” IEEE Trans. Evol. Comput., vol. 18, no. 3, pp. 450–455, Jun. 2014.
[37] K. Fang and D. Lin, “Uniform designs and their application in industry,” in Handbook on Statistics in Industry, vol. 22. Amsterdam, The Netherlands: Elsevier, 2003, pp. 131–170.
[38] M. Omidvar, X. Li, and X. Yao, “Smart use of computational resources based on contribution for cooperative co-evolutionary algorithms,” in Proc. Genet. Evol. Comput. Conf., Dublin, Ireland, 2011, pp. 1115–1122.
[39] R. Storn and K. Price, “Differential evolution—A simple and efficient heuristic for global optimization over continuous spaces,” J. Global Optim., vol. 11, no. 4, pp. 341–359, 1997.
[40] K. Deb, S. Agrawal, A. Pratap, and T. Meyarivan, “A fast and elitist multiobjective genetic algorithm: NSGA-II,” IEEE Trans. Evol. Comput., vol. 6, no. 2, pp. 182–197, Apr. 2002.
[41] L. Tseng and C. Chen, “Multiple trajectory search for unconstrained/ constrained multi-objective optimization,” in Proc. IEEE Congr. Evol. Comput., Trondheim, Norway, 2009, pp. 1951–1958.
[42] P. Bosman and D. Thierens, “The naive MIDEA: A baseline multi-objective EA,” in Proc. Evol. Multi-Criterion Optim. (EMO), Guanajuato, Mexico, 2005, pp. 428–442.
[43] C. Ahn, “Advances in evolutionary algorithms,” in Theory, Design and Practice. Berlin, Germany: Springer, 2006.
[44] Q. Zhang, W. Liu, and H. Li, “The performance of a new version of MOEA/D on CEC09 unconstrained MOP test instances,” School Comput. Sci. Electr. Eng., Univ. Essex, Colchester, U.K., Tech. Rep. 20091, 2009.
[45] M. Liu, X. Zou, Y. Chen, and Z. Wu, “Performance assessment of DMOEA-DD with CEC 2009 MOEA competition test instances,” in Proc. IEEE Congr. Evol. Comput., Trondheim, Norway, 2009, pp. 2913–2918.
[46] E. Zitzler, L. Thiele, M. Laumanns, C. Fonseca, and V. Fonseca, “Performance assessment of multiobjective optimizers: An analysis and review,” IEEE Trans. Evol. Comput., vol. 7, no. 2, pp. 117–132, Apr. 2003.
[47] J. Derrac, S. Garca, D. Molina, and F. Herrera, “A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms,” Swarm Evol. Comput., vol. 1, no. 1, pp. 3–18, 2011.
[48] J. Durillo et al., “A study of multiobjective metaheuristics when solving parameter scalable problems,” IEEE Trans. Evol. Comput., vol. 14, no. 4, pp. 618–635, Aug. 2010.
[49] L. Marti, J. Garcia, A. Berlanga, and J. Molina, “Introducing MONEDA: Scalable multiobjective optimization with a neural estimation of distribution algorithm,” in Proc. Genet. Evol. Comput. Conf. (GECCO), Atlanta, GA, USA, 2008, pp. 689–696.
[50] L. Marti, J. Garcia, A. Berianga, and J. Molina, “Multi-objective optimization with an adaptive resonance theory-based estimation of distribution algorithm,” Ann. Math. Artif. Intell., vol. 68, no. 4, pp. 247–273, 2013.