计算机科学 ›› 2023, Vol. 50 ›› Issue (10): 203-213.doi: 10.11896/jsjkx.220900242
马慧, 冯翔, 虞慧群
MA Hui, FENG Xiang, YU Huiqun
摘要： 进化多任务优化是计算智能领域一个新兴的研究方向,它致力于研究通过进化算法如何同时、有效地求解多个优化问题,从而提高单独求解每个任务的性能。基于此,提出了一种基于两层知识迁移的多代理多任务优化算法(AMS-MTO),其通过在代理间和代理内同时进行知识迁移来达到跨域优化的目的。具体来讲,代理内的知识迁移是通过差分进化实现决策变量信息的跨维迁移,从而避免算法陷入局部最优;代理间的学习采用了隐式知识迁移和显式知识迁移两种策略。隐式知识迁移利用种群的选择性交叉来产生后代,促进遗传信息的交流;显式知识迁移是对精英个体的迁移,可以弥补隐式迁移随机性很强的缺点。为了评估两层知识迁移的多代理多任务优化方法的有效性,在8个高达100维的基准问题上进行了实证研究,同时给出了收敛证明,并将其与现有的算法进行了对比。实验结果表明,在求解单目标优化的昂贵问题时,AMS-MTO算法效率更高,性能更好,收敛速度更快。
中图分类号:
| [1]DING J L,YANG C,JIN Y C,et al.Generalized multitasking for evolutionary optimization of expensive problems[J].IEEE Transactions on Evolutionary Computation,2019,23(1):44-58. [2]THEIL H.A rank-invariant method of linear and polynomial re-gression analysis[J].Advanced Studies in Theoretical and Applied Econometrics,1992,23:345-381. [3]KOURAKOS G,MANTOGLOU A.Pumping optimization ofcoastal aquifers based on evolutionary algorithms and surrogate modular neural network models[J].Advances in Water Resources,2009,32(4):507-521. [4]BUCHE D,SCHRAUDOLPH N,KOUMOUTSAKOS P.Acce-lerating evolutionary algorithms with gaussian process fitness function models[J].IEEE Transactions on Systems,Man and Cybernetics Part C:Applications and Reviews,2005,35(2):183-194. [5]GONZALEZ J,ROJAS I,ORTEGA J,et al.Multiobjective evolutionary optimization of the size,shape,and position parameters of radial basis function networks for function approximation[J].IEEE Transactions on Neural Networks,2003,14(6):1478-1495. [6]JIN Y C.Surrogate-assisted evolutionary computation:Recentadvances and future challenges[J].Swarm and Evolutionary Computation,2011,1(2):61-70. [7]HAFTKA R T,VILLANUEVA D,CHAUDHURI A.Parallel surrogate-assisted global optimization with expensive functions－－a survey[J].Structural and Multidisciplinary Optimization,2016,54:3-13. [8]VINCENZI L,GAMBARELLI P.A proper infill sampling stra-tegy for improving the speed performance of a surrogate-assisted evolutionary algorithm[J].Computers and Structures,2017,178:58-70. [9]TIAN J,TAN Y,ZENG J C,et al.Multiobjective infill criterion driven gaussian process-assisted particle swarm optimization of high-dimensional expensive problems[J].IEEE Transactions on Evolutionary Computation,2019,23(3):459-472. [10]GOH C K,LIM D,MA L,et al.A surrogate-assisted memeticco-evolutionary algorithm for expensive constrained optimization problems[C]//2011 IEEE Congress of Evolutionary Computation(CEC).2011:744-749. [11]LE M N,ONG Y S,MENZEL S,et al.Evolution by adapting surrogates[J].Evolutionary Computation,2013,21(2):313-340. [12]YU H B,TAN Y,SUN C L,et al.A generation-based optimal restart strategy for surrogate-assisted social learning particle swarm optimization[J].Knowledge-Based Systems,2019,163:14-25. [13]LI F,CAI X W,GAO L,et al.A surrogate-assisted multiswarm optimization algorithm for high-dimensional computationally expensive problems[J].IEEE Transactions on Cybernetics,2021,51(3):1390-1402. [14]ZHOU Z Z,ONG Y S,NAIR P.Hierarchical surrogate-assisted evolutionary optimization framework[C]//Proceedings of the 2004 Congress on Evolutionary Computation.2004:1586-1593. [15]LIM D,JIN Y C,ONG Y S,et al.Generalizing surrogate-assisted evolutionary computation[J].IEEE Transactions on Evolutionary Computation,2010,14(3):329-355. [16]SUN C L,JIN Y C,ZENG J C,et al.A two-layer surrogate-assisted particle swarm optimization algorithm[J].Soft Computing,2015,19:1461-1475. [17]YU H B,TAN Y,ZENG J C,et al.Surrogate-assisted hierarchical particle swarm optimization[J].Information Sciences,2018,454-455:59-72. [18]BALI K K,GUPTA A,FENG L,et al.Linearized domain adaptation in evolutionary multitasking[C]//2017 IEEE Congress on Evolutionary Computation(CEC).2017:1295-1302. [19]WEN Y W,TING C K.Parting ways and reallocating resources in evolutionary multitasking[C]//2017 IEEE Congress on Evolutionary Computation(CEC).2017:2404-2411. [20]LIAW R T,TING C K.Evolutionary many-tasking based onbiocoenosis through symbiosis:A framework and benchmark problems[C]//2017 IEEE Congress on Evolutionary Computation(CEC).2017:2266-2273. [21]LI G H,ZHANG Q F,GAO W F.Multipopulation evolution framework for multifactorial optimization[C]//Genetic and Evolutionary Computation Conference.Association for Computing Machinery,2018:215-216. [22]MIN A T W,ONG Y S,GUPTA A,et al.Multiproblem surrogates:Transfer evolutionary multiobjective optimization of computationally expensive problems[J].IEEE Transactions on Evolutionary Computation,2019,23(1):15-28. [23]GUPTA A,ONG Y S.Genetic transfer or population diversification? deciphering the secret ingredients of evolutionary multitask optimization[C]//2016 IEEE Symposium Series on Computational Intelligence(SSCI).2016:1-7. [24]KATTAN A,GALVAN E.Evolving radial basis function networks via GP for estimating fitness values using surrogate mo-dels[C]//2012 IEEE Congress on Evolutionary Computation.2012:1-7. [25]WILD S M,SHOEMAKER C A.Global convergence of radial basis function trust region derivative-free algorithms[J].SIAM Journal of Optimization,2011,21(3):761-781. [26]WILD S M,SHOEMAKER C A.Global convergence of radial basis function trust region algorithms for derivative-free optimization[J].SIAM Rev,2013,55(2):349-371. [27]DING J L,YANG C,JIN Y C,et al.Generalized multitasking for evolutionary optimization of expensive problems[J].IEEE Transactions on Evolutionary Computation,2019,23(1):44-58. [28]LIAO P,SUN C L,ZHANG G C,et al.Multi-surrogate multi-tasking optimization of expensive problems[J].Knowledge-Based Systems,2021,551:23-38. |
|
||
首页