A strict minimax inequality criterion and some of its consequences

TL;DR

This paper introduces a flexible scheme for strict minimax inequalities and explores its implications, including conditions for unique local minima in finite-dimensional Hilbert spaces with applications to convex and differentiable functions.

Contribution

It presents a novel, adaptable criterion for strict minimax inequalities and demonstrates its wide-ranging consequences in optimization and analysis.

Findings

Established a new minimax inequality criterion.

Proved existence and uniqueness of local minima under certain conditions.

Demonstrated applications to convex functions in finite-dimensional spaces.

Abstract

In this paper, we point out a very flexible scheme within which a strict minimax inequality occurs. We then show the fruitfulness of this approach presenting a series of various consequences. Here is one of them: Let $Y$ be a finite-dimensional real Hilbert space, $J:Y\to {\bf R}$ a $C^1$ function with locally Lipschitzian derivative, and $\varphi:Y\to [0,+\infty[$ a $C^1$ convex function with locally Lipschitzian derivative at 0 and $\varphi^{-1}(0)=\{0\}$. Then, for each $x_0\in Y$ for wich $J'(x_0)\neq 0$, there exists $\delta>0$ such that, for each $r\in ]0,\delta[$, the restriction of $J$ to $B(x_0,r)$ has a unique global minimum $u_r$ which satisfies $$J(u_r)\leq J(x)-\varphi(x-u_r)$$ for all $x\in B(x_0,r)$, where $B(x_0,r)=\{x\in Y: \|x-x_0\|\leq r\}\ .$