search

IOHprofiler / IOHexperimenter

Experimentation procedure for Iterative Optimization Heuristics
Other
47 stars 24 forks source link

Star Discrepancy optimum value #168

Closed dani2221 closed 1 year ago

dani2221 commented 1 year ago

when calling problem.optimum on any star discrepancy the output looks like this:

image

Which is obviously not the right answer. Tested on Star Discrepancy Integer and Star Discrepancy Real and encountered the problem on both.

borisalmonacid commented 1 year ago

y: inf is the fitness value???

jacobdenobel commented 1 year ago

Hi,

For star discrepancy problems the optimum is unknown. While we should be able to compute this for simpler problems in lower dimensions, this becomes infeasible for problems of higher dimensionality/number of points. Therefore, we decided not to provide any information on the optimum for any star discrepancy problems, which is why we only return the value (0, 0), inf for the location of the optimum, which basically means the optimum is unknown.

And yes, y is the location of the (unknown) optimum.