0

For a function $g: X \times Y \rightarrow \mathbb R$ show that:

$$\sup_{y\in Y}\inf_{x\in X} g(x,y) \leq \inf_{x\in X}\sup_{y\in Y} g(x,y)$$

My attempt:

By definition: $\inf_{\bar x\in X} g(\bar x,y) \leq \sup_{\bar y\in Y} g(x,\bar y)$ for each $x,y$. In particular, this is true for the pair $(x,y)$ such that $y$ maximizes the LHS and $x$ minimizes the RHS. So the claim holds.

My difficulty

In my proof, I've assumed that the infimum and supremum are attained. This isn't necessarily the case if $X,Y$ are not closed or if the infimum/supremum is at $\infty$. What is a way to clean up this argument? My thought is to instead consider sequences that approach infimum and supremum, respectively. Is there some more direct way?

dmh
  • 2,958
  • I would do this: $$\forall x,y:\ g(x,y) \leq \sup_{y^\in Y} g(x,y^)\ \implies \forall y:\ \inf_{x^\in X}g(x^,y)\le\inf_{x^\in X}\sup_{y^\in Y}g(x^,y^)\ \implies \sup_{y^\in Y}\inf_{x^\in X}g(x^,y^)\le\inf_{x^\in X}\sup_{y^\in Y}g(x^,y^)$$ – user Mar 13 '21 at 17:56
  • Why was this closed? I'm not asking for a solution, im asking for guidance and tips. – dmh Mar 13 '21 at 19:54

0 Answers0