# Unconstrained optimization

Taken literally, this is an unconstrained mathematical program. However, this phrase is also used in a context that $LaTeX: X$ could contain the strict interior, with constraints of the form $LaTeX: g(x) < 0,$ but the mathematical program behaves as unconstrained. This arises in the context of some algorithm design, as the solution is known to lie in the interior of $LaTeX: X,$ such as with the barrier function. As long as the algorithm has a continuous trajectory, this abuse is ok, but the constraints must be considered if one could jump, as in pattern search.