tidu.m

基于梯度训练方法的rbf神经网络实现

function main()
SamNum=100;
TargetSamNum=101;
InDim=1;
UnitNum=10;
MaxEpoch=5000;
E0=0.05;


rand('state',sum(100*clock))
NoidseVar=0.1;
Noise=NoidseVar*rand(1,SamNum);
SamIn=8*rand(1,SamNum)-4;
SamOutNoNoise=1.1*(1-SamIn+2*SamIn.^2).*exp(-SamIn.^2/2);
SamOut=SamOutNoNoise+Noise;

TargetIn=-4:0.08:4;
TargetOut=1.1*(1-TargetIn+2*TargetIn.^2).*exp(-TargetIn.^2/2);



Center=8*rand(InDim,UnitNum)-4;
SP=0.2*rand(1,UnitNum)+0.1;
W=0.2*rand(1,UnitNum)-0.1;
lrCent=0.01;
lrSP=0.001;
lrW=0.001;
Alpha=0.05;
ErrHistory=[];
for epoch=1:MaxEpoch
    AllDist=dist(Center',SamIn);
    SPMat=repmat(SP',1,SamNum);
    UnitOut=radbas(AllDist./SPMat);
    
    NetOut=W*UnitOut;
    Error=SamOut-NetOut;
    
    SSE=sumsqr(Error)
    
    ErrHistory=[ErrHistory SSE];
    
    if SSE<E0,break,end
    WGrad0=0;
    for i=1:UnitNum
        CentGrad=(SamIn-repmat(Center(:,i),1,SamNum))*(Error.*UnitOut(i,:)*W(i)/(SP(i)^2))';
        SPGrad=AllDist(i,:).^2*(Error.*UnitOut(i,:)*W(i)/(SP(i)^3))';
        WGrad=Error*UnitOut(i,:)';
        Center(:,i)=Center(:,i)+lrCent*CentGrad;
        SP(i)=SP(i)+lrSP*SPGrad;
        W(i)=W(i)+lrW*WGrad+Alpha*lrW*WGrad0;
        WGrad0=WGrad;
    end
end

TestDistance=dist(Center',TargetIn);
TestSpreadsMat=repmat(SP',1,TargetSamNum);
TestHiddenUnitOut=radbas(TestDistance./TestSpreadsMat);
TestNNOut=W*TestHiddenUnitOut;
plot(TargetIn,TargetOut,'k--',TargetIn,TestNNOut,'r+');

figure
hold on 
grid
[xx,Num]=size(ErrHistory);
plot(1:Num,ErrHistory,'r--');