一级黄色片免费播放|中国黄色视频播放片|日本三级a|可以直接考播黄片影视免费一级毛片

高級搜索

留言板

尊敬的讀者、作者、審稿人, 關(guān)于本刊的投稿、審稿、編輯和出版的任何問題, 您可以本頁添加留言。我們將盡快給您答復(fù)。謝謝您的支持!

姓名
郵箱
手機號碼
標(biāo)題
留言內(nèi)容
驗證碼

基于進化規(guī)劃的多層前向網(wǎng)絡(luò)結(jié)構(gòu)優(yōu)化

李映,  白本督,  焦李成

李映, 白本督, 焦李成. 基于進化規(guī)劃的多層前向網(wǎng)絡(luò)結(jié)構(gòu)優(yōu)化[J]. 電子與信息學(xué)報, 2001, 23(12): 1298-1302.
引用本文: 李映, 白本督, 焦李成. 基于進化規(guī)劃的多層前向網(wǎng)絡(luò)結(jié)構(gòu)優(yōu)化[J]. 電子與信息學(xué)報, 2001, 23(12): 1298-1302.
Li Ying, Bai Bendu, Jiao Licheng. STRUCTURAL OPTIMIZATION OF MULTILAYER FEEDFORWARD NETWORKS BASED ON EVOLUTIONARY PROGRAMMING[J]. Journal of Electronics & Information Technology, 2001, 23(12): 1298-1302.
Citation: Li Ying, Bai Bendu, Jiao Licheng. STRUCTURAL OPTIMIZATION OF MULTILAYER FEEDFORWARD NETWORKS BASED ON EVOLUTIONARY PROGRAMMING[J]. Journal of Electronics & Information Technology, 2001, 23(12): 1298-1302.

基于進化規(guī)劃的多層前向網(wǎng)絡(luò)結(jié)構(gòu)優(yōu)化

STRUCTURAL OPTIMIZATION OF MULTILAYER FEEDFORWARD NETWORKS BASED ON EVOLUTIONARY PROGRAMMING

  • 摘要: 基于進化規(guī)劃(EP)方法,該文提出了設(shè)計多層前向網(wǎng)絡(luò)拓?fù)浣Y(jié)構(gòu)和權(quán)值分布的一種新算法EPANN算法。EPANN算法能同時進化網(wǎng)絡(luò)的結(jié)構(gòu)和連接權(quán)值(包括閾值),在進化過程中,強調(diào)父代和子代之間的行為聯(lián)結(jié),結(jié)構(gòu)變異既有結(jié)點刪除,又有結(jié)點增加,不同于單純的刪除算法或構(gòu)造算法,且結(jié)點刪除總是先于結(jié)點增加,保證了網(wǎng)絡(luò)規(guī)模盡可能小而泛化能力盡可能強。
    Abstract: Based on evolutionary programming, a novel algorithm named EPANN for designing the topology and weight distributions of feedforward networks is proposed in this paper. EPANN algorithm evolves network architectures and connection weights (including biases) simultaneously and emphasizes the behavioral links between parents and their offspring in evolution, such as weights training after each architectural mutation and node splitting. Unlike the pure constructive or pruning algorithm, EPANN's architectural mutations include both node deletion and node addition, and prefer node deletion to addition in order to encourage the network architecture as compact as possible and generalization ability as good as possible.
    • HTML全文
  • R. Reed, Pruning algorithms A survey, IEEE Trans. on Neural Networks, 1993, NN-4(5), 740747.[2]T.Y. Kwok, D. Y. Yeung, Constructive algorithms for structure learning in feedforward neural networks for regression problems, IEEE Trans. on Neural Networks, 1997, NN-8(5), 1131 1148.[3]V. Maniezzo, Genetic evolution of the topology and weight distribution of neural networks, IEEE Trans. on Neural Networks, 1994, NN-5(1), 39-53.[4]P.J. Angeline, G. M. Saunders, J. B. Pollack, An evolutionary algorithm that constructs recurrent neural networks, IEEE Trans. on Neural Networks, 1994, NN-5(1), 54-64.[5]X. Yao.[J].Y. Liu, Evolutionary design of artificial neural networks with different nodes, Proc. of ICEC96. Nagoya, Japan.1996,:-[6]X. Yao. Y. Liu, A new evolutionary system for evolving artificial neural networks, IEEE Trans.on Neural Networks, 1997, NN-8(3), 694-713.[7]G.W. Greenwood. Training partially recurrent neural networks using evolutionary strategies IEEE Trans. on Speech Audio Processing, 1997. 5(1), 192-194.[8]P.G. Harrald. M. Kamstra, Evolving artificial neural networks to combine financial forecasts.IEEE Trans. on Evolutionary Computation, 1997, 1(1), 40-51.[9]徐立新,等,神經(jīng)網(wǎng)絡(luò)訓(xùn)練的一種隨機方法,哈爾濱工業(yè)大學(xué)報,1996,29(3),85-87[10]S.V. Odri. D. P. Petrovacki. G. A. Krstonosic. Evolutional development of a multilevel neural network. Neural Networks, 1993,6(4), 694-713.
  • 加載中
計量
  • 文章訪問數(shù):  1959
  • HTML全文瀏覽量:  98
  • PDF下載量:  406
  • 被引次數(shù): 0
出版歷程
  • 收稿日期:  1999-11-30
  • 修回日期:  2000-06-30
  • 刊出日期:  2001-12-19

目錄

    /

    返回文章
    返回