Problem with unconstrained optimization fminunc 'quasi-newton'
1 view (last 30 days)
Show older comments
I tried to find the minimum of a function with the algorithm 'quasi-newton', but a message appeared to me. It says
"Local minimum possible. fminunc stopped because it cannot decrease the objective function along the current search direction."
The followed path is detailed step-by-step below:
1)
function out =G(x)
a=x(1);
b=x(2);
out=(2*a.^2+8*b.^2-4*a)./exp(a.^2+b.^2-2*a-1);
end
2)x0=[1,0];
3)opts = optimoptions(@fminunc, 'Algorithm', 'quasi-newton');
4)[xopt,fopt]=fminunc(@G,x0,opts);
%It should work, because it seems to be rigorously done following all the rules. Nevertheless I suppose there is something wrong.
iterations: 1
funcCount: 12
stepsize: []
lssteplength: []
firstorderopt: 1.0729e-06
algorithm: 'quasi-newton'
message: 'Local minimum possible.…
Thank you very much in advance
0 Comments
Answers (1)
Alan Weiss
on 16 Nov 2015
If you want to examine the quality of the solution, see the documentation on "local minimum possible".
Alan Weiss
MATLAB mathematical toolbox documentation
0 Comments
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!