最近,彭在文献[1]中提出了关于拟半E-凸函数的一个判别准则。本文首先利用E-凸函数和凸函数的定义给出了E-凸函数的一个等价条件,即在E:R^n→R^n,M真包含R^n是一个E-凸集,E(M)是凸集,f是定义在M上的实值函数的情况下,若函数f在M上是E-凸的当且仅当φ(λ)=f[E(y)+λ(E(x)-E(y))]在[0,1]上是凸函数。其次,本文对文献[1]中关于拟半E-凸函数的结论进行了研究分析,指出其结论在本质上来说可以退化到拟凸函数的情形。
Recently a new criterion of quasi-semi-E-convex functions was introduced by Peng in 2006 for a new criterion of quisi-semi- E-convex functions. In this paper, firstly, we propose a necessary and sufficient condition for E-convex functions by using the definitions of E-convex function and convex function, that is, Let E:R^n→R^n,M lohtain in R^n is a E-convex set, E(M) is a convex set, f is a realvalued function defined in M, so f is E-convex function if and only if φ( λ ) =f[ E(y) + λ (E(x) -E(y))] is convex function under some conditions. Then we have the other theorem, let f: M→R is a upper semi-continuous function in convex set M lohtain in R^n, if there exists a linear mapping E: R^n→R^n, such that E(M) lohtain in M, for A↓ x∈ M have f(x) ≤f( E (x)) , and yn→y as E( yn )→E( y), then f is a quasi-semi-E-convex function←→ β∈ (0,1) (or β =0,1 ), such that f(βE(x) + (1 -β)E(y) ) ≤max{f(x) ,f(y)} , A↓ ,y∈M.