0
$\begingroup$

I am interested in determining whether the function $F:L^{\infty}(0,\infty;L^{\infty}(0,1)) \to \mathbb R$ defined by $$u \mapsto F(u)=\int_0^\infty \int_0^1 u(x,t)^2 \ \mathrm dx \, \mathrm dt $$ is differentiable/Frechét differentiable at an element $u \in L^2(0,\infty;L^2(0,1))$.

Any suggestions or assistance would be greatly appreciated.

$\endgroup$
0

1 Answer 1

4
$\begingroup$

First of all, instead of $L^{\infty}(0,\infty;L^{\infty}(0,1))$, you should write $L^{\infty}([0,\infty);L^{\infty}(0,1))$ (or maybe $L^{\infty}((0,\infty);L^{\infty}(0,1))$, depending of what you want here).

Second, for any $u\in L^{\infty}([0,\infty);L^{\infty}(0,1))$ (say), any $t\in[0,\infty)$, and any $x\in(0,1)$, the value at $x$ of the value $u(t)$ of $u$ at $t$ should be written as $u(t)(x)$, rather than $u(x,t)$.

Finally, there is no function $F$ from $L^{\infty}([0,\infty);L^{\infty}(0,1))$ (or from $L^{\infty}((0,\infty);L^{\infty}(0,1))$) to $\mathbb R$ such that $$F(u)=\int_0^{\infty} \int_0^1 u(t)(x)^2 \,dx \,dt$$ for all $u\in L^{\infty}([0,\infty);L^{\infty}(0,1))$. Indeed, if $u(t)(x)=1$ for all $t\in[0,\infty)$ and $x\in(0,1)$, then $u\in L^{\infty}([0,\infty);L^{\infty}(0,1))$ but $\int_0^{\infty} \int_0^1 u(t)(x)^2 \,dx \,dt=\infty$. (The case of $L^{\infty}((0,\infty);L^{\infty}(0,1))$ is quite similar.)

So, your question is about a nonexistent function $F$, and hence any answer about $F$ would be correct.

$\endgroup$
6
  • $\begingroup$ Maybe the OP means $\overline{\mathbb R}$ as a codomain of $F$… ? the whole question is unclear $\endgroup$ Commented Dec 10, 2023 at 15:42
  • 5
    $\begingroup$ I like so much the conclusion $\endgroup$ Commented Dec 10, 2023 at 15:43
  • $\begingroup$ @PietroMajer : Thank you for your comments. Concerning your first comment: If a function $F$ takes an infinite value at some point $u$ (and also at some point in every neighborhood of $u$), how would we define a derivative of $F$ at $u$? I think we would then have to deal with something like $\infty-\infty$. $\endgroup$ Commented Dec 10, 2023 at 16:33
  • $\begingroup$ I don’t know… maybe the question was: prove that it is differentiable at u in the domain of F if u is also in $L^2$ etc (then I guess the answer would be no, it is not even continuous wrto the $L^\infty$ topology) $\endgroup$ Commented Dec 10, 2023 at 17:57
  • $\begingroup$ @PietroMajer You are correct; this is what I wanted to understand. As you mentioned, the map is not even continuous in the general case. $\endgroup$
    – elmas
    Commented Dec 11, 2023 at 13:34

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .