Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

In Hilbert space why does convergence in weak topology $x_n$ to $x$ imply that $x_n$ converges to $x$ in norm?

Thank you very much for your answers. What if I put a condition on weak convergence i.e., suppose it also holds $\lim_{n\to \infty } \lVert x_n\rVert \to \lVert x\rVert$ then can I say that $x_n$ converges in norm?

I doubt if it always holds!

I think to understand this I need bit more explanation because I am struggling with the understanding of weak topologies. Thank you

share|cite|improve this question
up vote 10 down vote accepted

$\def\p#1{\left\langle#1\right\rangle}$That is not true. Consider $\ell^2$ with the standard inner product $$ \p{x,y} = \sum_{i=0}^\infty x_i\bar y_i $$ and $e_n = (0,\ldots, 0, 1, 0,\ldots)\in \ell^2$ (1 at position $n$). Then for $x \in \ell^2$ we have $$ \p{e_n, x} = \bar x_n \to 0 = \p{0, x}$$ so $e_n \to 0$ weakly. But $\|e_n\| = 1 \not \to 0$, so $e_n \not\to 0$ in norm.

What is true in every Hilbert space is that $x_n \to x$ weakly and $\|x_n\| \to \|x\|$ implies $\|x_n - x\|\to 0$, that is $x_n \to x$ in norm. To see that, write \begin{align*} \|x_n - x\|^2 &= \|x_n\|^2 - 2\Re\p{x_n, x} + \|x\|^2 \end{align*} Now the first term converges to $\|x\|^2$ by assumption and the second one to $2\Re\p{x,x} = 2\|x\|^2$ by weak convergence, so alltogether $$ \|x_n -x\|^2 = \|x_n\|^2 - 2\Re\p{x_n, x}+ \|x\|^2 \to \|x\|^2 - 2\|x\|^2 + \|x\|^2 = 0 $$

share|cite|improve this answer
:Thanks a lot. What about if i add my new condition on the question. – Theorem Dec 2 '12 at 20:16
Have you actually read martini's answer? – Chris Eagle Dec 2 '12 at 20:20
It might be worth noting that OPs original claim does hold in $\ell^1(\mathbb N)$ although it is not a Hilbert space. – JSchlather Dec 2 '12 at 21:29

This is false if the space is infinite-dimensional. For example, let $e_n$ be an orthonormal sequence. Then $e_n \to 0$ weakly, but not in norm.

share|cite|improve this answer
what would change if i add extra property like above . – Theorem Dec 2 '12 at 20:17

It is not true in general, as already mentioned, but it is true with the added condition. Suppose that $x_n\to x$ weakly, and that $\|x_n\|\to\|x\|$. Then $$ \|x-x_n\|^2=\langle x-x_n,x-x_n\rangle = \|x\|^2+\|x_n\|^2-2\text{Re}\,\langle x_n,x\rangle\to\|x\|^2+\|x\|^2-2\|x\|^2=0. $$

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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