Initial point is a local minimum that satisfies the constraints

UQ with UQLab Community Q&A and How To
1

Dear UQLabcommunity,
I’m using the Kriging metamodel in UQLab, but I face a problem when using it. The matlab shows the following message.

"— Calculating the Kriging metamodel… —
Optimization terminated: average change in the fitness value less than options.FunctionTolerance.

Initial point is a local minimum that satisfies the constraints.

Optimization completed because at the initial point, the objective function is non-decreasing
in feasible directions to within the selected value of the optimality tolerance, and
constraints are satisfied to within the default value of the constraint tolerance.

"

Could anybody tell me what’s wrong with my codes?
I think it seems that some setup may be not right, but I did not find where to modify the setup of my metamodel.

2

Hello @GPLai2 and welcome to UQWorld!

Which optimization method are you using?

Generally speaking this is not a bug but rather an expected behavior when using a local (gradient-based) optimization method and the objective function is flat around the starting point.
Did you also notice poor performance of the resulting surrogate?

If you could provide us the code that reproduces this behavior we could give you more detailed feedback.