4
$\begingroup$

Let $V$ be the Banach space of bounded sequences of reals with the sup norm. Does there exists a subset $B$ of $V$ such that

  • Linear Independence: For all functions $c$ in $\mathbb{R}^B$, if $\sum_{b \in B} c(b) \cdot b = 0$, then $c$ is identically zero.
  • Spanning Set: For all vectors $v$ in $V$, there exists a function $c$ in $\mathbb{R}^B$ such that $\sum_{b \in B} c(b) \cdot b = v$.

If so, is an explicit such $B$ known?

Improve this question
$\endgroup$
6
  • $\begingroup$ he Banach space V with which you start would seem to be $\ell^\infty_{\mathbb R}$ and not $L^\infty({\mathbb R})$ -- as it happens, the two are (non-isometrically) isomorphic as Banach spaces, but this is not trivial and the isomorphism is slightly mysterious. Could you clarify whether you had one or the other in mind when you asked this question? $\endgroup$
    – Yemon Choi
    Aug 9, 2010 at 4:50
  • $\begingroup$ Why the set of $e_j=(0,\ldots,0,1,0\ldots)$ does not work ? $\endgroup$
    – Leandro
    Aug 9, 2010 at 4:51
  • $\begingroup$ Yemon, I had the sequence space in mind. Leandro, that's not a spanning set. $\endgroup$
    – user5810
    Aug 9, 2010 at 4:55
  • $\begingroup$ Are you requiring these functions to be zero for all but finitely many $b$? If so, this is just the statement that every vector space has a basis. I think you can't find an explicit basis. $\endgroup$
    – Kevin Ventullo
    Aug 9, 2010 at 5:24
  • $\begingroup$ No, and that's why I made my question more explicit than the topic. $\endgroup$
    – user5810
    Aug 9, 2010 at 5:38

2 Answers 2

Reset to default
10
$\begingroup$

The space $\ell^\infty_R$ does not have even an M-basis; i.e., a biorthogonal set $(x_t,x_t^*)$ such that the span of the $x_t$ is dense and the $x_t^*$ are total (Lindenstrauss, late 1960s IIRC), so it has nothing like a Schauder basis. Later I proved [PAMS 26. no. 3 467-468 (1970)] that $\ell^\infty$ also does not have an M-basis. However, each of these spaces does have a biorthogonal set $(x_t,x_t^*)$ such that the span of the $x_t$ is dense. This is in my paper with W.J. Davis [Studia Math. 45 173-179 (1973)].

Improve this answer
$\endgroup$
2
  • $\begingroup$ What does "the $x_t^*$ are total" mean? $\endgroup$
    – user5810
    Aug 9, 2010 at 14:53
  • 1
    $\begingroup$ Separate points. $\endgroup$
    – Bill Johnson
    Aug 9, 2010 at 15:03
0
$\begingroup$

It exists for any linear space (Banach structure is not essential here), is called Hamel base. No explicit construction (without use of Axiom of Choice) exists (and, I guess, it may proved in a sense - that existence of Hamel base in any linear space implies the axiom of choice or smth similar).

Improve this answer
$\endgroup$
8
  • 1
    $\begingroup$ No, a Hamel base is what you would get if you required the cs to be zero for all but finitely many b. $\endgroup$
    – user5810
    Aug 9, 2010 at 5:59
  • $\begingroup$ If $c$ is not assumed to be finitely supported, then how do you understand this sum? Should it be absolutely convergent series in almost every point? $\endgroup$
    – Fedor Petrov
    Aug 9, 2010 at 6:09
  • $\begingroup$ See en.wikipedia.org/wiki/… $\endgroup$
    – user5810
    Aug 9, 2010 at 6:12
  • $\begingroup$ So, $B$ is not countable, but $c$ is countably supported and the convergence is unconditional in $L_{\infty}$ norm. Right? $\endgroup$
    – Fedor Petrov
    Aug 9, 2010 at 6:20
  • 1
    $\begingroup$ Ricky: regarding the Wikipedia page, my mistake. I still think the question might be a little clearer if you said explicitly that you were talking about unconditional convergemce of the "series" - especially to indicate that you are talking about something a little different from a Schauder basis $\endgroup$
    – Yemon Choi
    Aug 9, 2010 at 8:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.