You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
26 lines
803 B
Matlab
26 lines
803 B
Matlab
function [y,Y,nmse]=dim_activation_profile(W,x,y,iterations,V)
|
|
[n,m]=size(W);
|
|
psi=1;%5000; %need to learn using e-(1/psi)
|
|
epsilon1=0.0001; %>0.001 this becomes significant compared to y and hence
|
|
%produces sustained responses and more general suppression
|
|
epsilon2=100*epsilon1*psi;%this determines scaling of initial transient response
|
|
|
|
if nargin<3 | isempty(y), y=zeros(n,1,'single'); end
|
|
|
|
if nargin<4, iterations=25; end
|
|
if nargin<5,
|
|
%set feedback weights equal to feedforward weights normalized by maximum value
|
|
%V=W./(1e-9+(max(W')'*ones(1,m)));
|
|
V=bsxfun(@rdivide,W,(1e-9+max(W,[],2)));
|
|
U=V;
|
|
end
|
|
|
|
for i=1:iterations
|
|
%update responses
|
|
e=x./(epsilon2+(V'*y));
|
|
y=(epsilon1+y).*(W*e);
|
|
|
|
Y(:,i)=y;
|
|
end
|
|
%euclid=sqrt(sum((x-V'*y).^2));
|
|
nmse=sum((x-V'*y).^2)./sum(x.^2); |