进化神经网络间接编码方法的计算框架研究

doi:10.3969/j.issn.0372-2112.2013.05.004

电子学报 ›› 2013, Vol. 41 ›› Issue (5): 852-858.DOI: 10.3969/j.issn.0372-2112.2013.05.004

进化神经网络间接编码方法的计算框架研究

蔺想红, 李志强, 魏伟一

西北师范大学计算机科学与工程学院,甘肃兰州 730070

收稿日期:2012-05-29 修回日期:2013-01-08 出版日期:2013-05-25
- 作者简介:
- 蔺想红 男,1976年1月生于甘肃天水.西北师范大学计算机科学与工程学院副教授,博士,硕士生导师.研究方向为神经网络、进化计算、人工生命、图像处理. E-mail:linxh@nwnu.edu.cn
- 基金资助:
- 国家自然科学基金 (No.61165002); 甘肃省自然科学基金 (No.1010RJZA019); 西北师范大学科研基金 (No.NWNU-LKQN-10-3)

A Computational Framework for Indirect Encoding Methods of Evolutionary Neural Networks

LIN Xiang-hong, LI Zhi-qiang, WEI Wei-yi

School of Computer Science and Engineering,Northwest Normal University, Lanzhou,Gansu 730070,China

Received:2012-05-29 Revised:2013-01-08 Online:2013-05-25 Published:2013-05-25

摘要/Abstract

摘要： 针对传统的直接编码方法对大规模神经网络难以进化的问题,研究者提出了进化神经网络的间接编码方法,这类方法的核心思想是网络子结构的重复可通过一组基因的多次表达来实现从基因型到表现型的映射,这种基因重用机制可在较小的基因空间中进行大规模神经网络的快速搜索.本文在总结和分析各类间接编码实现方法的基础上,给出了进化神经网络间接编码方法的一般性计算框架,每一次神经网络的进化过程分为三个阶段:发育阶段、学习阶段和进化阶段.并从计算框架的基因组进化过程和神经网络发育过程两个方面比较分析了各种典型间接编码方法的优缺点.

关键词: 进化神经网络, 间接编码, 发育控制, 网络结构

Abstract: According to the difficulties in the evolving large scale neural networks using the traditional direct encoding methods,many researchers are proposing the novel indirect encoding methods for evolutionary neural networks.That is,a network structure that repeats many times can be represented by a single set of genes that is reused in mapping from genotype to phenotype,and such genetic reuse allows searching the large scale neural networks through a lower dimensional genotypic space.In this paper,we introduce a general computational framework for the indirect encoding methods of evolutionary neural networks through the study of existed methods,in which every evolutionary process of neural networks is divided into three stages:development,learning and evolution.Additionally,we analyze the advantages and disadvantages for the different indirect encoding methods from two aspects of the computational framework:genome evolution and neural network development.

Key words: evolutionary neural network, indirect encoding, developmental control, network structure

中图分类号:

TP183

蔺想红, 李志强, 魏伟一. 进化神经网络间接编码方法的计算框架研究[J]. 电子学报, 2013, 41(5): 852-858.

LIN Xiang-hong, LI Zhi-qiang, WEI Wei-yi. A Computational Framework for Indirect Encoding Methods of Evolutionary Neural Networks[J]. Acta Electronica Sinica, 2013, 41(5): 852-858.

参考文献

[1] Schaffer J D,Whitley D,Eshelman L J.Combinations of genetic algorithms and neural networks:A survey of the state of the art .Proceedings of International Workshop on Combinations of Genetic Algorithms and Neural Networks .Los Alamitos,CA:IEEE Computer Society Press,1992.1-37.
[2] Yao X.A review of evolutionary artificial neural networks[J].International Journal of Intelligent Systems,1993,8(4):539-567.
[3] Whitley D.An overview of evolutionary algorithms:Practical issues and common pitfalls[J].Information and Software Technology,2001,43(14):817-831.
[4] Yao X.Evolving artificial neural networks[J].Proceedings of the IEEE,1999,87(9):1423-1447.
[5] Stanley K O,Miikkulainen R.A taxonomy for artificial embryogeny[J].Artifical Life,2003,9(2):93-130.
[6] Roggen D,Federici D,Floreano D.Evolutionary morphogenesis for multi-cellular systems[J].Genetic Programming and Evolvable Machines,2007,8(1):61-96.
[7] Clune J,Stanley K O,Pennock R T,Ofria C.On the performance of indirect encoding across the continuum of regularity[J].IEEE Transactions on Evolutionary Computation,2011,15(3):346-367.
[8] Kitano H.Designing neural networks using genetic algorithms with graph generation system[J].Complex Systems,1990,4(4):461-476.
[9] Boers E J W,Kuiper H.Biological metaphors and the design of modular artificial neural networks .The Netherlands:Leiden University,1992.
[10] Campos L M L,Roisenberg M,Oliveira R C L.Automatic design of neural networks with L-systems and genetic algorithms-A biologically inspired methodology .The 2011 International Joint Conference on Neural Networks .Piscataway,NJ:IEEE Press,2011.1199-1206.
[11] Jung S Y.A topographical method for the development of neural networks for artificial brain evolution[J].Artificial Life,2005,11(3):293-316.
[12] Mouret J B,Doncieux S.Mennag:A modular,regular and hierarchical encoding for neural-networks based on attribute grammars[J].Evolutionary Intelligence,2008,1(3):187-207.
[13] Gruau F.Neural network synthesis using cellular encoding and the genetic algorithm .France:Ecole Normale Suprieure de Lyon,1994.
[14] Luke S,Spector L.Evolving graphs and networks with edge encoding:Preliminary report .Late Breaking Papers at the Genetic Programming Conference .Stanford,CA:Stanford Bookstore,1996.117-124.
[15] Suchorzewski M.Evolving scalable and modular adaptive networks with developmental symbolic encoding[J].Evolutionary Intelligence,2011,4(3):145-163.
[16] Astor J C,Adami C.A developmental model for the evolution of artificial neural networks[J].Artificial Life,2000,6(3):189-218.
[17] Dellaert F,Beer R.D.A developmental model for the evolution of complete autonomous agents .Proceedings of the 4th International Conference on Simulation of Adaptive Behavior .Cambridge,MA:MIT Press,1996.393-401.
[18] Federici D.A regenerating spiking neural network[J].Neural Networks,2005,18(56):746-754.
[19] 蔺想红,张田文,张贵仓.进化大规模脉冲神经网络的发育方法[J].计算机学报,2012,35(12):2633-2644. Lin Xianghong,Zhang Tianwen,Zhang Guicang.A developmental method for evolving large-scale spiking neural networks[J].Chinese Journal of Computers,2012,35(12):2633-2644.(in Chinese)
[20] Stanley K O.Compositional pattern producing networks:A novel abstraction of development[J].Genetic Programming and Evolvable Machines,2007,8(2):131-162.
[21] Stanley K O,D'Ambrosio D B,Gauci J.A hypercube-based encoding for evolving large-scale neural networks[J].Artificial Life,2009,15(2):185-212.
[22] Risi S,Stanley K O.An enhanced hypercube-based encoding for evolving the placement,density,and connectivity of neurons[J].Artificial Life,2012,18(4):331-363.
[23] Hoang T H,McKay R I,Essam D,Hoai N X.On synergistic interactions between evolution,development and layered learning[J].IEEE Transactions on Evolutionary Computation,2011,15(3):287-312.
[24] Lin X H,Zhang T W.Structure and dynamics of artificial regulatory networks evolved by segmental duplication and divergence model[J].International Journal of Automation and Computing,2010,7(1):105-114.
[25] Stanley K O,Miikkulainen R.Evolving neural networks through augmenting topologies[J].Evolutionary Computation,2002,10(2):99-127.

进化神经网络间接编码方法的计算框架研究

A Computational Framework for Indirect Encoding Methods of Evolutionary Neural Networks

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 6

编辑推荐

Metrics

[1]	周涛, 霍兵强, 陆惠玲, 任海玲. 残差神经网络及其在医学图像处理中的应用研究[J]. 电子学报, 2020, 48(7): 1436-1447.
[2]	谭翔元, 高晓光, 贺楚超. 基于马尔科夫毯约束的最优贝叶斯网络结构学习算法[J]. 电子学报, 2019, 47(9): 1898-1904.
[3]	潘剑飞, 董一鸿, 陈华辉, 钱江波, 戴明洋. 基于结构紧密性的重叠社区发现算法[J]. 电子学报, 2019, 47(1): 145-152.
[4]	唐雨, 何明升, 王英赫. 基于用户偏好的嵌入性网络结构分析[J]. 电子学报, 2012, 40(7): 1339-1344.
[5]	朱近康. 未来移动通信的技术挑战和解决方案[J]. 电子学报, 2004, 32(S1): 6-10.
[6]	刘凯;李建东. 分布式无线网络中的自适应控制协议[J]. 电子学报, 2001, 29(7): 877-880.