Higher regularity of homeomorphisms in the Hartman-Grobman theorem and a conjecture on its sharpness

TL;DR

This paper proves that in the Hartman-Grobman theorem, the conjugating homeomorphism can be Lipschitz continuous but not differentiable, with an inverse that is H"older continuous but not Lipschitz, and discusses the sharpness of these regularity results.

Contribution

It establishes the first result showing the homeomorphism's Lipschitz regularity in the Hartman-Grobman theorem and introduces a conjecture on the optimality of this regularity.

Findings

Homeomorphism is Lipschitz but not $C^1$.

Inverse homeomorphism is H"older but not Lipschitz.

Examples demonstrate the regularity properties and support the conjecture.

Abstract

Hartman-Grobman theorem states that there is a homeomorphism H sending the solutions of the nonlinear system onto those of its linearization under suitable assumptions. Many mathematicians have made contributions to prove H\"older continuity of the homeomorphisms. However, is it possible to improve the H\"older continuity to Lipschitzian continuity? This paper gives a positive answer. We formulate the first result that the homeomorphism is Lipschitzian, but not $C^1$, while its inverse is merely H\"{o}lder continuous, but not Lipschitzian. It is interesting that the regularity of the homeomorphism is different from its inverse. Moreover, some illustrative examples are presented to show the effectiveness of our results. Further, motivated by our example, we also propose a conjecture, saying, the regularity of the homeomorphisms is sharp and it could not be improved any more.