Please wait a minute...
FITEE
Front. Inform. Technol. Electron. Eng.  2016, Vol. 17 Issue (4): 283-295    DOI: 10.1631/FITEE.1500306
    
自适应分布式搜索过程中的组织变化及问题复杂度对性能的影响
Friederike Wall
Department of Controlling and Strategic Management, Alpen-Adria-Universitaet Klagenfurt, 9020 Klagenfurt, Austria
Organizational dynamics in adaptive distributed search processes: effects on performance and the role of complexity
Friederike Wall
Department of Controlling and Strategic Management, Alpen-Adria-Universitaet Klagenfurt, 9020 Klagenfurt, Austria
 全文: PDF 
摘要： 目的：研究组织设置变更对分布式自适应搜索过程的影响，特别关注局部搜索间的交互复杂度。
创新点：基于分布式多智能体的仿真分析了组织设置的不同模式变化对于不同复杂度的搜索问题的影响。
方法：首先介绍了所采用的智能体仿真模型，描述了构建在NK适应度曲面上的分布式搜索问题及其复杂度。接着阐述了组织设置包括搜索智能体和核心智能体的设置、搜索智能体的视角形成、不同搜索智能体之间的协调以及搜索多样性等。然后给出组织动态性的数学描述，并进行了实验分析：（1）比较了组织变化对子问题交互复杂度最大和最小两种极端搜索问题的影响（称之为“基准”）；（2）分析最终和平均优化性能对搜索问题复杂度的敏感性；（3）分析交叉智能体交互复杂度对性能影响的不稳定性。
结论：本文的研究表明组织动态变化能够增加分布式自适应搜索的有效性，比如提高最终性能表现、增加获取最优解的可能性以及发现新的解域等。这种有效性增加幅度很大程度上取决于组织变化模式以及问题的复杂度。
关键词： 基于智能体的仿真;  复杂度;  协调;  分布式搜索;  NK曲面    
Abstract: In this paper, the effects of altering the organizational setting of distributed adaptive search processes in the course of search are investigated. We put particular emphasis on the complexity of interactions between partial search problems assigned to search agents. Employing an agent-based simulation based on the framework of NK landscapes we analyze different temporal change modes of the organizational set-up. The organizational properties under change include, for example, the coordination mechanisms among search agents. Results suggest that inducing organizational dynamics has the potential to increase the effectiveness of distributed adaptive search processes with respect to various performance measures like the final performance achieved at the end of the search, the chance to find the optimal solution of the search problem, or the average performance per period achieved during the search process. However, results also indicate that the mode of temporal change in conjunction with the complexity of the search problem considerably affects the order of magnitude of these beneficial effects. In particular, results suggest that organizational dynamics induces a shift towards more exploration, i.e., discovery of new areas in the fitness landscape, and less exploitation, i.e., stepwise improvement.
Key words: Agent-based simulation    Complexity    Coordination    Distributed search    NK landscapes
收稿日期: 2015-09-21 出版日期: 2016-04-05
CLC:  TP13  
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　
Friederike Wall

引用本文:

Friederike Wall. Organizational dynamics in adaptive distributed search processes: effects on performance and the role of complexity. Front. Inform. Technol. Electron. Eng., 2016, 17(4): 283-295.

链接本文:

http://www.zjujournals.com/xueshu/fitee/CN/10.1631/FITEE.1500306    或    http://www.zjujournals.com/xueshu/fitee/CN/Y2016/V17/I4/283

[1] Guo-jiang Shen, Yong-yao Yang. 一种城市主干道信号动态协调控制方法及其应用[J]. Front. Inform. Technol. Electron. Eng., 2016, 17(9): 907-918.
[2] Shui-qing Gong, Jing Chen, Qiao-yan Kang, Qing-wei Meng, Qing-chao Zhu, Si-yi Zhao. 面向虚拟SDN网络的高效协调映射算法[J]. Front. Inform. Technol. Electron. Eng., 2016, 17(7): 701-716.
[3] Yu-Tang Zhu, Jun-yong Liu, Yong-Bo Zhao, Jun Liu, Peng-Lang Shui. 基于正定约束广义秩信号模型的低复杂度稳健自适应算法[J]. Front. Inform. Technol. Electron. Eng., 2016, 17(11): 1245-1252.
[4] Michaelraj Kingston Roberts, Ramesh Jayabalan. 一种改进的用于低密度奇偶校验码的低复杂度和积译码[J]. Front. Inform. Technol. Electron. Eng., 2015, 16(6): 511-518.
[5] Zhi-min Han, Zhi-yun Lin, Min-yue Fu, Zhi-yong Chen. 图拉普拉斯视角下的多智能体系统分布式协调控制[J]. Front. Inform. Technol. Electron. Eng., 2015, 16(6): 429-448.