3
$\begingroup$

I want to show that the following operator is compact:

$$T:\mathbb \ell^p\rightarrow \mathbb \ell^p, \text{ }(x_n)_n\mapsto(\frac{x_n}{n})_n \text{ } 1\leq p<\infty$$

Its the first time that I am trying to show that an operator is compact.

I know the following three definitions of a compact operator:

Let $T:X\rightarrow Y$ be a bounded operator, then $T$ is compact if

1) The image of the unit ball is relatively compact or

2) The image of any bounded set in X is relatively compact or

3) Any bounded sequence $(x_n)$ in $X$ has a subsequence such that $Kx_{n_k}$ converges.

But I feel like none of this definitions can help me to prove the compactness of the operator directly.

Is there something like a "general way" to show this? At least for Operators from $\mathbb \ell^p\rightarrow \ell^p$ ?

Thanks in advance

$\endgroup$
2
  • $\begingroup$ The general way that I got used to (at least for relatively nice operators) is using the third formulation. In other words, take a bounded sequence and show there is in fact a convergent subsequence. $\endgroup$
    – DanZimm
    Commented Jan 17, 2015 at 20:26
  • $\begingroup$ You can use the ideas in (or just the result of) this post to show $T(B(\ell_p))$ is totally bounded (and thus relatively compact). You might also find this helpful. $\endgroup$ Commented Jan 17, 2015 at 20:42

1 Answer 1

15
+100
$\begingroup$

Hint: If an operator can be approximated by finite rank operators, then it is compact. Try to show that $\|T-T_k\| \to 0$ for a suitable chosen sequence $\{ T_k\}$ of finite rank operators.

Try with $$ T_k: (x_1, x_2, \ldots)\mapsto (x_1, \frac{x_2}{2}, \ldots, \frac{x_k}{k}, 0, 0, \ldots ). $$ It is obvious that $rk(T_k)=k$, i.e., it is finite rank operator. Since $$ \| T-T_k\|^p=\sup\{ \sum_{n=k+1}^{\infty}\left|\frac{x_n}{n}\right|^p;\quad \sum_{n=1}^{\infty}|x_n|^p\leq 1\}\leq $$ $$\leq \frac{1}{(k+1)^p}\sup\{ \sum_{n=k+1}^{\infty}|x_n|^p;\quad \sum_{n=1}^{\infty}|x_n|^p\leq 1\} \leq \frac{1}{(k+1)^p}$$ we have $$ \| T-T_k\|\leq\frac{1}{k+1} \to 0\quad (k\to \infty). $$

$\endgroup$
4
  • $\begingroup$ Should't it be a $C<\infty$ instead of 1 for the bound? $\endgroup$
    – Duke
    Commented Jan 22, 2015 at 0:09
  • $\begingroup$ @Duke $\sum_{n=k+1}^{\infty}|x_n|^p \leq \sum_{n=1}^{\infty}|x_n|^p\leq 1$. $\endgroup$ Commented Jan 22, 2015 at 5:43
  • $\begingroup$ Excuse me but I mean $\sum_{n=1}^{\infty}|x_n|^p\leq 1$. I just want to know if there is a reason why you took 1. $\endgroup$
    – Duke
    Commented Jan 22, 2015 at 7:01
  • $\begingroup$ @Duke The definition of the operator norm: $\| A\|=\sup\{ \| Ax\|; \| x\|\leq 1\}$. $\endgroup$ Commented Jan 22, 2015 at 7:09

You must log in to answer this question.

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