3
$\begingroup$

It is not difficult to show that the sequence space $\ell_1$ is a subspace of $\ell_2$ that is not closed in the $\ell_2$ topology (see Problem 2 in this pdf file for instance)

What is the closure of $\ell_1$ in $\ell_2$?

$\endgroup$
6
$\begingroup$

$\ell^1 $ is dense in $\ell^2$ (w.r.t. the $\ell^2$ norm topology). The argument is simple: given $\textbf{x} = (x_1, \ldots x_n, \ldots) \in\ell^2$ fix $\epsilon >0$ and take $\textbf{x}_N = (x_1, \ldots x_N, 0, 0, 0 \ldots)$ where $N$ is such that $\displaystyle \left(\sum_{k=N+1}^\infty x_k^2\right)^{1/2} < \epsilon.$

Then $\textbf{x}_N$ is obviously in $\ell^1$ and $\|\textbf{x} - \textbf{x}_N\|_2 < \epsilon$.

Hence the closure of $\ell^1$ in the $\ell^2$ topology is $\ell^2$.

$\endgroup$
4
$\begingroup$

All $e_n$ are in $\ell_1$. What is the closure of their span in $\ell_2$?

$\endgroup$
0
$\begingroup$

Closure of $l_p$ in $l_q$ is $l_q$ for any $p,q$ with $p<q$. Or, $l_p$ is dense in $l_q$ for any $p,q$ with $p<q$. You can prove it either by using the concept/arguement given by Henno Brandsma (above) or by using the technic given by David Bowman (above).

$\endgroup$

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.