Weakly Convergent l^p Sequences: Boundedness and Convergence in C

  • Context: Graduate 
  • Thread starter Thread starter e12514
  • Start date Start date
Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
1 reply · 2K views
e12514
Messages
27
Reaction score
0
For 1 < p < oo it is true that:
If we have a sequence ( x_n )_(n >= 1) in l^p that converges weakly to zero then that implies the x_n are uniformly bounded and that ((x^{m})_n) -> 0 in C (as n -> oo) for each fixed m.
(where l^p is the space of p-summable sequences of complex numbers, and I wrote x_n = ((x^{m})_n )_(m >= 1) , couldn't think of a better notation...)

Is the converse true? Why would that be? It doesn't look too obvious... at least not for me.
 
Physics news on Phys.org


Provided I've read your post correctly, then yes, the converse is true.

So basically you want to show that [itex]|f(x_n) - f(0)| \to 0[/itex] for all [itex]f \in (\ell^p)^\ast = \ell^q[/itex] given pointwise convergence and uniform boundedness. First consider the space [itex]\ell[/itex] of finite sequences. For every [itex]f \in \ell^\ast[/itex], we have that [itex]\lim_n f(x_n) = 0[/itex] (why?). Now use the fact that [itex]\ell[/itex] is dense in all the [itex]\ell^p[/itex] spaces.