IEEE有关神经网络基于MATLAB的论文文献.rar

  • PUDN用户
    了解作者
  • matlab
    开发工具
  • 1.3MB
    文件大小
  • rar
    文件格式
  • 0
    收藏次数
  • 10 积分
    下载积分
  • 7
    下载次数
  • 2019-07-16 21:21
    上传日期
MATLAB simulink 径向基神经网络 前馈神经网络
IEEE有关神经网络基于MATLAB的论文文献.rar
  • 06010199Application of BP Neural Network to determine of mine water inrush sources based on Matlab.pdf
    174.6KB
  • 05458945Application and Comparison of BP Neural Network Algorithm in MATLAB.pdf
    288.3KB
  • 05212498EXTENDING VERSION OF GRAPHICAL USER INTERFACE IN NEURAL.pdf
    988.6KB
  • 05287701PID Ramp Controller Regulated by Fuzzy RBF Neural Network.pdf
    614.2KB
  • 04633804基于MATLAB的在线时变Sylvester方程解的张子网仿真建模与仿真.pdf
    192KB
内容介绍
<html xmlns="http://www.w3.org/1999/xhtml"> <head> <meta charset="utf-8"> <meta name="generator" content="pdf2htmlEX"> <meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1"> <link rel="stylesheet" href="https://static.pudn.com/base/css/base.min.css"> <link rel="stylesheet" href="https://static.pudn.com/base/css/fancy.min.css"> <link rel="stylesheet" href="https://static.pudn.com/prod/directory_preview_static/625004556caf596192f98250/raw.css"> <script src="https://static.pudn.com/base/js/compatibility.min.js"></script> <script src="https://static.pudn.com/base/js/pdf2htmlEX.min.js"></script> <script> try{ pdf2htmlEX.defaultViewer = new pdf2htmlEX.Viewer({}); }catch(e){} </script> <title></title> </head> <body> <div id="sidebar" style="display: none"> <div id="outline"> </div> </div> <div id="pf1" class="pf w0 h0" data-page-no="1"><div class="pc pc1 w0 h0"><img class="bi x0 y0 w1 h1" alt="" src="https://static.pudn.com/prod/directory_preview_static/625004556caf596192f98250/bg1.jpg"><div class="t m0 x1 h2 y1 ff1 fs0 fc0 sc0 ls0 ws0">Proceedings of<span class="_ _0"> </span> the Eighth International Conference on Ma<span class="ls1 ws1">chine Learning and Cybernetics<span class="_ _1"></span>, Baoding, 12-15 J<span class="_ _1"></span>uly 2009 </span></div><div class="t m0 x2 h3 y2 ff2 fs0 fc0 sc0 ls2 ws2">978-1-4244-3703-0/09/$25.00 &#169;2009 IE<span class="_ _1"></span>EE </div><div class="t m0 x3 h3 y3 ff2 fs0 fc0 sc0 ls3 ws3">187 </div><div class="t m0 x4 h4 y4 ff3 fs1 fc0 sc0 ls4 ws4">EXTENDING<span class="_ _2"></span> VERSION OF<span class="_ _2"></span> <span class="_ _3"></span>GRAPHICA<span class="ls5 ws5">L<span class="_ _2"></span> USER INTERF<span class="_ _4"></span>ACE IN NEURAL<span class="_ _2"></span> </span></div><div class="t m0 x5 h4 y5 ff3 fs1 fc0 sc0 ls6 ws6">NETWORK T<span class="_ _2"></span>OOLBOX OF<span class="_ _2"></span> MA<span class="_ _4"></span>TLAB <span class="_ _2"></span>A<span class="ls7 ws7">ND ENGINEERING<span class="_ _2"></span> <span class="_ _5"></span>APPLICA<span class="_ _4"></span>TIONS </span></div><div class="t m0 x6 h5 y6 ff3 fs0 fc0 sc0 ls8 ws8">FANG-YUAN XU, LONG ZHOU, YING-N<span class="_ _3"></span>AN MA, LOI LEI LAI </div><div class="t m0 x7 h6 y7 ff4 fs0 fc0 sc0 ls9 ws9">City Univer<span class="_ _1"></span>sity Londo<span class="_ _1"></span>n, Northam<span class="_ _2"></span>pton Square, London, <span class="_ _1"></span>United Ki<span class="_ _1"></span>ngdom<span class="_ _1"></span> <span class="_"> </span> </div><div class="t m0 x8 h6 y8 ff4 fs0 fc0 sc0 lsa wsa">E-MAIL: datuan12345@hotmail.com, long.zho<span class="_ _3"></span>u.1@city<span class="_ _5"></span><span class="lsb wsb">.ac.uk, y<span class="_ _5"></span>.ma@city<span class="_ _5"></span>.ac.uk<span class="_ _3"></span>, l.l.lai<span class="_ _1"></span>@city<span class="_ _5"></span>.ac.uk </span></div><div class="t m0 x2 h7 y9 ff3 fs2 fc0 sc0 lsc ws3">Abstract: </div><div class="t m0 x9 h7 ya ff3 fs2 fc0 sc0 lsd wsc">This paper pr<span class="_ _1"></span>oposes an extending version of Graphica<span class="_ _1"></span>l </div><div class="t m0 x2 h7 yb ff3 fs2 fc0 sc0 lse wsd">User Interface in Neural Network T<span class="_ _4"></span>oolbox of MA<span class="_ _2"></span>TLAB 7.1 </div><div class="t m0 x2 h7 yc ff3 fs2 fc0 sc0 lsf wse">which releas<span class="_ _1"></span>es the limit of <span class="ls10 wsf">setting mor<span class="_ _2"></span>e layers in the </span></div><div class="t m0 x2 h7 yd ff3 fs2 fc0 sc0 ls11 ws10">feedforward network <span class="_ _1"></span>creating<span class="_ _1"></span>. Users can set u<span class="_ _1"></span>p a feedforward </div><div class="t m0 x2 h7 ye ff3 fs2 fc0 sc0 lse ws11">network with any architectur<span class="_ _2"></span>e<span class="_ _3"></span>. <span class="ls12 ws12">Based on this interface <span class="_ _1"></span>two </span></div><div class="t m0 x2 h7 yf ff3 fs2 fc0 sc0 ls13 ws13">simple applications are applied, name<span class="_ _1"></span>ly<span class="_ _2"></span>, a simple voting </div><div class="t m0 x2 h7 y10 ff3 fs2 fc0 sc0 ls14 ws14">system design and 3&#8211;phase generator output detector design. </div><div class="t m0 x2 h7 y11 ff3 fs2 fc0 sc0 ls15 ws15">Simple voting system is based on feedforw<span class="_ _3"></span>ard network and the </div><div class="t m0 x2 h7 y12 ff3 fs2 fc0 sc0 ls14 ws16">3&#8211;phase generator output detector<span class="lse ws17"> is designed on Radial Basis </span></div><div class="t m0 x2 h7 y13 ff3 fs2 fc0 sc0 ls16 ws18">Network (RBF). Thes<span class="_ _1"></span>e examples sh<span class="_ _1"></span>ow how to design<span class="_ _1"></span> the </div><div class="t m0 x2 h7 y14 ff3 fs2 fc0 sc0 ls17 ws19">neural network in applications. </div><div class="t m0 x2 h5 y15 ff3 fs0 fc0 sc0 ls18 ws1a">Keywords: </div><div class="t m0 xa h5 y16 ff3 fs0 fc0 sc0 ls19 ws1b">Artificial neural networ<span class="ws1c">k; Pl<span class="_ _1"></span>atform; Neurons; </span></div><div class="t m0 x2 h6 y17 ff3 fs0 fc0 sc0 ls1a ws1d">Learning rate;Pea<span class="_ _1"></span>k value; Feedforward<span class="_ _1"></span> network; RBF<span class="ff4 ls1b ws3"> </span></div><div class="t m0 x2 h3 y18 ff3 fs0 fc0 sc0 ls1c ws3">1.<span class="ff2 ls1b"> <span class="_ _6"> </span></span><span class="ls1d">Introduction </span></div><div class="t m0 xa h6 y19 ff4 fs0 fc0 sc0 ls1e ws1e">Artificial neural network is a model abstracted from </div><div class="t m0 x2 h6 y1a ff4 fs0 fc0 sc0 ls1f ws1f">biological neural<span class="_ _1"></span> network. In biol<span class="_ _2"></span>ogical neural network, </div><div class="t m0 x2 h6 y1b ff4 fs0 fc0 sc0 ls1b ws20">&#8220;when a neuron recei<span class="_ _1"></span>ves an excitatory input which is lar<span class="_ _2"></span>g<span class="_ _3"></span>e<span class="_ _1"></span> </div><div class="t m0 x2 h6 y1c ff4 fs0 fc0 sc0 ls20 ws21">enough compared with its inh<span class="_ _3"></span>ibitory input, it will sends a </div><div class="t m0 x2 h6 y1d ff4 fs0 fc0 sc0 ls21 ws22">spike of electrical activity down its axon. Learning occ<span class="_ _1"></span>urs </div><div class="t m0 x2 h6 y1e ff4 fs0 fc0 sc0 ls19 ws23">by changing t<span class="_ _1"></span>he ef<span class="_ _1"></span>fectiveness of the sy<span class="_ _2"></span>napses so that the </div><div class="t m0 x2 h6 y1f ff4 fs0 fc0 sc0 ls22 ws24">influence of<span class="_ _1"></span> one neur<span class="_ _1"></span>on on anot<span class="_ _1"></span>her changes&#8221;<span class="_ _1"></span>[1]. <span class="_"> </span> </div><div class="t m0 xa h6 y20 ff4 fs0 fc0 sc0 ls18 ws25">Comparing wi<span class="_ _1"></span>th other m<span class="_ _1"></span>odels, the m<span class="_ _1"></span>ain advantage o<span class="_ _1"></span>f </div><div class="t m0 x2 h6 y21 ff4 fs0 fc0 sc0 ls23 ws26">artificial neural network is to be ab<span class="_ _3"></span>le to extract patterns and </div><div class="t m0 x2 h6 y22 ff4 fs0 fc0 sc0 ls24 ws27">detect trends that are too complex to be noticed <span class="_ _1"></span>by either </div><div class="t m0 x2 h6 y23 ff4 fs0 fc0 sc0 ls25 ws28">humans or<span class="_ _1"></span> other c<span class="_ _1"></span>omputer<span class="_ _1"></span> techniques.<span class="_ _1"></span> Also, a<span class="_ _2"></span>rtificial neural </div><div class="t m0 x2 h6 y24 ff4 fs0 fc0 sc0 ls26 ws29">network has the ability of<span class="_ _3"></span> self-or<span class="_ _2"></span>gan<span class="_ _3"></span>ization, meaning that </div><div class="t m0 x2 h6 y25 ff4 fs0 fc0 sc0 ls27 ws2a">the system is intelligent. </div><div class="t m0 xa h6 y26 ff4 fs0 fc0 sc0 ls28 ws2b">Due to the splendid abilities of<span class="_ _3"></span> artificial neural </div><div class="t m0 x2 h6 y27 ff4 fs0 fc0 sc0 ls1d ws2c">network, the fa<span class="_ _1"></span>mous calcul<span class="_ _1"></span>ating software<span class="_ _1"></span>, MA<span class="_ _4"></span>TLAB, ma<span class="_ _1"></span>de </div><div class="t m0 x2 h6 y28 ff4 fs0 fc0 sc0 ls29 ws2d">a toolbox especially for artifi<span class="ls2a ws2e">cial<span class="_ _1"></span> neural network. T<span class="_ _1"></span>his </span></div><div class="t m0 x2 h6 y29 ff4 fs0 fc0 sc0 ls2b ws2f">toolbox gathers lo<span class="_ _3"></span>ts <span class="_ _1"></span>of functions for b<span class="_ _3"></span>a<span class="_ _1"></span>sic artificial neural </div><div class="t m0 x2 h6 y2a ff4 fs0 fc0 sc0 ls1 ws30">networks. Also MA<span class="_ _4"></span>TLAB has a Graphical <span class="_ _1"></span>User Interface of </div><div class="t m0 x2 h6 y2b ff4 fs0 fc0 sc0 ls22 ws31">neural netw<span class="_ _1"></span>ork for the i<span class="_ _1"></span>ntroducti<span class="_ _1"></span>on of neural<span class="_ _1"></span> network,<span class="_ _2"></span> </div><div class="t m0 x2 h6 y2c ff4 fs0 fc0 sc0 ls22 ws24">which can set<span class="_ _1"></span> at most 2 l<span class="_ _1"></span>ayers <span class="_ _1"></span>for feedfor<span class="_ _1"></span>ward networ<span class="_ _1"></span>k. </div><div class="t m0 xa h6 y2d ff4 fs0 fc0 sc0 ls18 ws32">In the view a<span class="_ _1"></span>bove, the m<span class="_ _1"></span>ain objective <span class="_ _1"></span>of this pape<span class="_ _1"></span>r is </div><div class="t m0 x2 h6 y2e ff4 fs0 fc0 sc0 lsb wsb">to introduce a new Graphical <span class="ls1 ws1">User Interface for feedforward </span></div><div class="t m0 xb h6 y2f ff4 fs0 fc0 sc0 ls2c ws33">network whic<span class="_ _2"></span>h can set as many layers as the use<span class="_ _1"></span>r wants. </div><div class="t m0 xb h6 y30 ff4 fs0 fc0 sc0 ls2d ws34">And based on this progr<span class="_ _3"></span>am<span class="_ _1"></span>, 2 applications ar<span class="_ _3"></span>e applied to </div><div class="t m0 xb h6 y31 ff4 fs0 fc0 sc0 ls2b ws35">introduce the way for designing<span class="_ _3"></span> artificial neural network. <span class="_"> </span> </div><div class="t m0 xb h3 y32 ff3 fs0 fc0 sc0 ls1c ws3">2.<span class="ff2 ls1b"> <span class="_ _6"> </span></span><span class="ls0 ws0">Extending graphical user interface <span class="_"> </span> </span></div><div class="t m0 xc h6 y33 ff4 fs0 fc0 sc0 ls0 ws36">The graphical user interface is<span class="ls2e ws37"> designed to be sim<span class="_ _2"></span>ple<span class="_ _3"></span> </span></div><div class="t m0 xb h6 y34 ff4 fs0 fc0 sc0 ls28 ws38">and user friendly<span class="_ _2"></span>. Users could type the data in the relative </div><div class="t m0 xb h6 y35 ff4 fs0 fc0 sc0 ls28 ws39">boxes with easy-understanding titles instead of a series of </div><div class="t m0 xb h6 y36 ff4 fs0 fc0 sc0 ls2b ws3a">codes. In the Neural Network T<span class="_ _2"></span>oolbox of MA<span class="_ _4"></span>TLA<span class="_ _3"></span>B <span class="_ _1"></span>7.1, </div><div class="t m0 xb h6 y37 ff4 fs0 fc0 sc0 ls2f ws3b">there is a graph<span class="_ _3"></span>ical user interface, &#8216;nntool&#8217;. But this </div><div class="t m0 xb h6 y38 ff4 fs0 fc0 sc0 ls1d ws3c">interface can <span class="_ _1"></span>only design<span class="_ _1"></span> a feedforwa<span class="_ _1"></span>rd network<span class="_ _1"></span> having <span class="_ _1"></span>2 </div><div class="t m0 xb h6 y39 ff4 fs0 fc0 sc0 ls30 ws3d">layers at most. Here a new interface, &#8216;platform&#8217;, is </div><div class="t m0 xb h6 y3a ff4 fs0 fc0 sc0 ls26 ws3e">introduced in Fig<span class="_ _3"></span>ure 1: </div><div class="t m0 xd h6 y3b ff4 fs0 fc0 sc0 ls1b ws3"> </div><div class="t m0 xe h7 y3c ff3 fs2 fc0 sc0 ls31 ws3f">Figure 1. The main menu of platform </div><div class="t m0 xc h6 y3d ff4 fs0 fc0 sc0 ls9 ws40">On this m<span class="_ _2"></span>ain platform, user<span class="_ _1"></span> could s<span class="_ _2"></span>elect buttons fo<span class="_ _1"></span>r </div><div class="t m0 xb h6 y3e ff4 fs0 fc0 sc0 ls27 ws41">their operations: </div><div class="t m0 xf h8 y3f ff5 fs0 fc0 sc0 ls1b ws3">&#8226;<span class="ff6"> <span class="_ _6"> </span><span class="ff4 ls32 ws42">&#8216;Data_creat&#8217;<span class="_ _5"></span> or &#8216;Network_c<span class="ls2 ws43">reat&#8217;: Create data or </span></span></span></div><div class="t m0 x10 h6 y40 ff4 fs0 fc0 sc0 ls33 ws44">neural network in this p<span class="_ _3"></span>l<span class="_ _1"></span><span class="ls24 ws45">atform. <span class="_ _1"></span>The created data </span></div><div class="t m0 x10 h6 y41 ff4 fs0 fc0 sc0 ls18 ws46">or network wi<span class="_ _1"></span>ll appear i<span class="_ _2"></span>n the corresponding <span class="_ _1"></span>list </div><div class="t m0 x10 h6 y42 ff4 fs0 fc0 sc0 ls34 ws3">box. </div></div><div class="pi" data-data='{"ctm":[1.611639,0.000000,0.000000,1.611639,0.000000,0.000000]}'></div></div> </body> </html>
评论
    相关推荐