how to write the code to smooth a function and then minimize it

Dear fellows,

I am a little confused with how to use the function smooth in matlab. At the moment I have a function f(x), here x contains 6 independent variable. I would like to first smooth this function then minimize it. So I write the code as follows.

if true
  % [X,F,exitflag] = ga(@(x)smooth(f(x,spf,DD,C),'rlowess'),6,[],[],Aeq,1,lb,ub,[],[]);
end

But ga still fails to find the correct minimum (which I know where it is as I am doing a simulation) and returns exact the same answer as in the case of just minimizing f. I am not sure where is this problem. Could you shed some light on this?

Thanks, Xueqi

Answer 1

[Shashank Prasanna]

GA is quite robust to discontinuities and 'non-smooth' objective functions. Did you give it a try without smoothing?

When the solver finishes, it exists under certain conditions. You may be able to relax/tighten tolerances for better results. I highly recommend these steps in the documentation:

When the Solver Fails

When the Solver Might Have Succeeded

When the Solver Succeeds