Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

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

Let $H:=\left\{u\in L^2(R^N):\nabla u \in L^2(R^N)\right\}$ and a functional $$f(u)=\int_{R^N} |\nabla u|^2dx+\left(\int_{R^N} |\nabla u|^2dx\right)^2.$$

If $\{u_n\}\subset H$ is a sequence such that $u_n \rightharpoonup u$, is it true that

$$\liminf_{n\rightarrow +\infty}\left[\int_{R^N} |\nabla u_n|^2dx+\left(\int_{R^N} |\nabla u_n|^2dx\right)^2\right]\geq \int_{R^N} |\nabla u|^2dx+\left(\int_{R^N} |\nabla u|^2dx\right)^2 \;?$$

share|cite|improve this question

Given any Hilbert space, if $u_n\rightharpoonup u$ weakly, then $\liminf_{n\to\infty}\|u_n\| \ge \|u\|$.

Proof: W.l.o.g. $\|u\| = 1$. $$ \|u\| = \langle u,u \rangle = \lim_{n\to\infty} \langle u,u_n \rangle $$ and $$ \langle u,u_n \rangle \le \|u_n\| $$ so $$ \lim_{n\to\infty} \langle u,u_n \rangle \le \liminf_{n\to\infty}\|u_n\|$$

Actually a similar argument works on any Banach space, and it is not just for Hilbert spaces.

share|cite|improve this answer

I guess the inner product is given by $$\langle u,v\rangle=\int_{\mathbb R^N}u(x)v(x)dx+\sum_{j=1}^N\int_{\mathbb R^N}\dfrac{\partial u}{\partial x_j}(x)\dfrac{\partial v}{\partial x_j}(x)dx.$$ Note that we can find a subsequence $\{u_{n_k}\}$ such that $\lim_{k\to\infty}f(u_{n_k})=\liminf_{n}f(u_n)$, and since the sequence $\{|\nabla u_{n_k}|\}$ and $\{u_{n_k}\}$ are bounded in $L^2(\mathbb R^N)$ , we can extract a subsequence such that $\{u_{\varphi(n)}\}$ and $\{\nabla u_{\varphi(n)}\}$ are weakly convergent in $L^2$. We denote by $v$ and $(v_1,\ldots,v_N)$ these limits. Then weak convergence shows that $\dfrac{\partial v}{\partial x_j}=v_j$, so $u=v$. We can conclude since in an Hilbert space, if a sequence $\{x_n\}$ converges weakly to $x$ then $\lVert x\rVert\leq \liminf_n \lVert x_n\rVert$: \begin{align*} \liminf_n\: f(u_n)&=\lim_{k\to \infty}f(u_{\varphi(k)})\\ &= \liminf_{k\to\infty}\:\left(\int_{\mathbb R^N}|\nabla u_{\varphi(k)}(x)|^2dx+\left(\int_{\mathbb R^N}|\nabla u_{\varphi(k)}(x)|^2dx\right)^2\right)\\ &\geq \liminf_{k\to\infty}\:\int_{\mathbb R^N}|\nabla u_{\varphi(k)}(x)|^2dx+ \left(\liminf_{k\to\infty}\int_{\mathbb R^N}|\nabla u_{\varphi(k)} (x)|^2dx\right)^2\\ &=\int_{\mathbb R^N}|\nabla u(x)|^2dx+\left(\int_{\mathbb R^N}|\nabla u(x)|^2dx\right)^2\\ &=f(u). \end{align*}

share|cite|improve this answer
Thank your answer, Dvaide Girudo! – Jiahua Jin Dec 2 '11 at 9:17
But I cannot see that $u_n\rightharpoonup u$ guarantees finding a subsequence $\{u_{n_k}\}$ such that $lim_{k\rightarrow+\infty} f(u_{n_k})=\liminf_{n\rightarrow} f(u_{n_k} )$. – Jiahua Jin Dec 2 '11 at 9:24
I have get your meaning, the limit of the weak convergance sequence is the unique and same as its any convergant subsequence. Your precise and detailed answer makes me clear. – Jiahua Jin Dec 2 '11 at 13:28
@Davide Giraudo: I would like to ask your comments in the following question… Thank you for your kind help – blindman Sep 4 '13 at 5:00

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.