scientific article; zbMATH DE number 2076406
In an earlier paper by the authors [Appl. Math. Lett. 13, No. 1, 27--30 (2000; Zbl 0952.26008)] it is proved that a function \(f: C\to\mathbb R\) defined on a nonempty convex set \(C\subseteq\mathbb R^n\) is convex if and only if it is quasiconvex and there exists some \(\alpha \in (0,1)\) such that \[ f(\alpha x+(1- \alpha)y)\leq \alpha f(x)+ (1-\alpha) f(y),\quad\forall x,y\in C. \] In the paper under review they show that the above characterization remains valid if one replaces the term ``quasiconvex with ``semistrictly quasiconvex. The two characterizations are independent from each other, since a semistrictly quasiconvex function need not be quasiconvex and viceversa. They also prove that one can weaken the semistrict quasiconvexity condition by replacing it with semilocal semistrict quasiconvexity, which means that for every \(x,y\in C\) such that \(f(x)>f(y)\) one has \(f(\alpha x+(1-\alpha )y)<f(x)\) for all sufficiently small \(\alpha >0.\) As the authors acknowledge at the end of the paper, some closely related characterizations of convexity were obtained earlier by \textit{F. A. Behringer} [Optimization 24, No. 3--4, 219--228 (1992; Zbl 0815.39009)].
- A note on convex functions
- A characterization of convex function
- A characterization of almost everywhere convex functions
- Convexity is equivalent to midpoint convexity combined with strict quasiconvextiy
- A characterization of convex functions
- A characterization of convex functions
- Characterizations of \(r\)-convex functions
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4469863)
