2
$\begingroup$

I saw an exercise today that says that : Let $f:\mathbb R\to \mathbb R$ a continuous and injective function. Show that the inverse $g:f(\mathbb R)\to \mathbb R$ is continuous. And under, there is a remark that says that it's not true in higher dimension. Could someone give me an example ? If $f:\mathbb R^n\to \mathbb R^n$ is injective and continuous, I know by Brouwer invariance domain that $f:\mathbb R^n\to f(\mathbb R^n)$ is a homeomorphism, so it should be a weirder function. Unfortunately, I can't imagine it. Could someone give me an example of such function ? May be it's a function of $$\mathbb R^n\supset \Omega \to f(\Omega )\subset \mathbb R^m,$$ but I don't think that there is bijective function if $n\neq m$. If such function doesn't exist in finite dimension, may be in infinite dimension it does ?

$\endgroup$
0
0
$\begingroup$

Consider the map $f:\Bbb R\to\Bbb R^2$, $$f(x)=\begin{cases}\left(\frac{1-x^2}{x^2+1},\frac{2x}{x^2+1}\right)&\text{if }x\ge 0\\ (x+1,0)&\text{if }x<0\end{cases}.$$ It parametrizes injectively the union of the ray $(-\infty,1]\times\{0\}$ with the upper half of the unit circle ($f(\Bbb R)$ is sort of a $P$-shaped set). Notice, however, that $f^{-1}:P\to\Bbb R^2$ cannot be continuous, because $f$ sends the non-compact set $[-2,\infty)$ onto the compact set $$\{(a,b)\in\Bbb R^2\,:\, (a^2+b^2=1\land b\ge 0)\lor(-1\le a\le 1\land b=0)\}.$$

This $f$ specifically turns out not to be differentiable, but it is entirely possible to parametrize a figure like the symbol "$\infty$" by a $C^\infty$ function $f:\Bbb R\to\Bbb R^2$ such that $f'$ has no zeros.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.