基於前期效能從多個粒子群優化演算法的設定模型中選擇最佳的設定模型

Selecting the Best Setting Model from Multiple Setting Models of Particle Swarm Optimization Based on the Previous Performance

黃俞典(Yu-Tien Huang)

指導教授 ： 張炎清

中山醫學大學/健康管理學院/醫學資訊學系碩士班/碩士(2017年)

摘要

在搜索目標函數最佳解的演算法領域中，粒子群優化演算法為簡單且有效的演算法。它的優點包括簡單且易於理解和實現，但其缺點是容易陷入區域最佳解中。為了保持原有的優點並提高其效能，我們在試著提出並驗證一個新穎的想法，這個構想乃是基於前期效能選擇最佳粒子群優化演算法的設定模型；設定模型的選擇是透過一種在具有多個設定模型的粒子群優化演算法中做切換。實驗結果證明，透過以此方案執行的粒子群優化演算法表現優於其它個別的模組設定。在未來，我們相信具備多模型選擇能力的粒子群優化演算法必將是富有前景的研究領域。此外，這個概念可以輕易地延伸到一種可從多個優化演算法中選擇最好的優化演算法的方案。

關鍵字

優化 ； 粒子群優化 ； 演算法 ； 切換

並列摘要

In the field of searching the optimal solutions of objective functions, particle swarm optimization (PSO) can be said to be a simple but effective algorithm. Its advantages include simplicity and ease to understand and implement, but it easily leads to getting stuck in local optima. In order to maintain the original benefit and promote its performance, we propose a novel idea in this paper, which selects the best setting model of PSO based on the previous performance through a switch of PSO with multiple setting models. Experimental results show that the PSO through the scheme is better than any with its individual setting alone. In the future, PSO algorithms with a switch of multiple models will be a promising research field. In addition, the idea can be easily extended to a scheme of selecting the best from multiple optimization methods.

並列關鍵字

optimization ； particle swarm optimization ； algorithm ； switch

參考文獻

[1]Proceedings of the IEEE International Conference on Neural Networks, pp. 1942–1948, December 1995.

[3]Y. Shi and R. Eberhart, “Parameter selection in particle swarm optimization,” Evolutionary Programming VII Lecture Notes in Computer Science, vol. 1447, pp 591-600, 1998.

[4]Y. Shi and R. Eberhart, “Empirical study of particle swarm optimization,” Proceedings of the 1999 Congress on Evolutionary Computation, pp. 1945-1950, 1999.

[5]M. Clerc and J. Kennedy, “The particle swarm: Explosion, stability, and convergence in a multi-dimensional complex space,” IEEE Trans. Evol. Comput., vol. 6, pp. 58–73, 2002.

[6]M. Clerc, “The swarm and the queen towards a deterministic and adaptive particle swarm optimization,” Proceedings of the 1999 Congress on Evolutionary Computation, pp. 1951-1957, 1999.