Revisiting Genetic Network Programming (GNP): Towards the Simplified Genetic Operators

被引:4
作者
Li, Xianneng [1 ]
Yang, Huiyan [1 ]
Yang, Meihua [1 ]
机构
[1] Dalian Univ Technol, Fac Management & Econ, Dalian 116024, Peoples R China
来源
IEEE ACCESS | 2018年 / 6卷
基金
中国国家自然科学基金;
关键词
Directed graph; evolutionary algorithms; genetic network programming; invalid/negative evolution; transition by necessity; OPTIMIZATION; ALGORITHM; EVOLUTION; MODEL;
D O I
10.1109/ACCESS.2018.2864253
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Genetic network programming (GNP) is a relatively new type of graph-based evolutionary algorithm, which designs a directed graph structure for its individual representation. A number of studies have demonstrated its expressive ability to model complicated problems/systems and explored it from the perspectives of methodologies and applications. However, the unique features of its directed graph are relatively unexplored, which cause unnecessary dilemma for the further usage and promotion. This paper is dedicated to uncover this issue systematically and theoretically. It is proved that the traditional GNP with uniform genetic operators does not consider the "transition by necessity'' feature of the directed graph, which brings the unnecessary difficulty of evolution to cause invalid/negative evolution problems. Consequently, simplified genetic operators are developed to address these problems. Experimental results on two benchmark testbeds of the agent control problems are carried out to demonstrate its superiority over the traditional GNP and the state-of-the-art algorithms in terms of fitness results, search speed, and computation time.
引用
收藏
页码:43274 / 43289
页数:16
相关论文
共 34 条
  • [1] [Anonymous], 1975, ADAPTATION NATURAL A
  • [2] [Anonymous], 2015, Reinforcement Learning: An Introduction
  • [3] [Anonymous], 2003, Genetic programming IV: routine human-competitive machine intelligence
  • [4] A real-coded genetic algorithm for training recurrent neural networks
    Blanco, A
    Delgado, M
    Pegalajar, MC
    [J]. NEURAL NETWORKS, 2001, 14 (01) : 93 - 105
  • [5] An algorithmic description of XCS
    M. V. Butz
    S. W. Wilson
    [J]. Soft Computing, 2002, 6 (3) : 144 - 153
  • [6] Butz M. V., 2000, 2000027 UIUC
  • [7] Gradient descent methods in learning classifier systems: Improving XCS performance in multistep problems
    Butz, MV
    Goldberg, DE
    Lanzi, PL
    [J]. IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2005, 9 (05) : 452 - 473
  • [8] A model of portfolio optimization using time adapting genetic network programming
    Chen, Yan
    Mabu, Shingo
    Hirasawa, Kotaro
    [J]. COMPUTERS & OPERATIONS RESEARCH, 2010, 37 (10) : 1697 - 1707
  • [9] Dollo's law and the re-evolution of shell coiling
    Collin, R
    Cipriani, R
    [J]. PROCEEDINGS OF THE ROYAL SOCIETY B-BIOLOGICAL SCIENCES, 2003, 270 (1533) : 2551 - 2555
  • [10] Differential Evolution: A Survey of the State-of-the-Art
    Das, Swagatam
    Suganthan, Ponnuthurai Nagaratnam
    [J]. IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2011, 15 (01) : 4 - 31