Quasiconjugates of functions, duality relationship between quasiconvex minimization under a reverse convex constraint and quasiconvex maximization under a convex constraint, and applications (Q811400)

The author introduces the following quasi-conjugate for the function f: \(R^ n\to \bar R:\) \(f^ H(v)=-\inf \{f(x)|\) \(<x,v>\geq 1\}\) if \(v\neq 0\), \(f^ H(v)=-\sup \{f(x)|\) \(x\in R^ n\}\) if \(v=0\). It is shown that \(f^ H\) is quasiconvex and if \(f(0)=\inf \{f(x)\}|\) \(x\in R^ n\setminus \{0\}\}\) then \(f^ H\) coincides with the quasiconvex hull of f, that \(f^ H\) is l.s.c. if f is u.s.c. and \(f^ H\) is u.s.c. if f is l.s.c. and achieves the maximum value at infinity. A duality theorem is also established. Some applications are then provided.