I am working through an example in "A Course on Functional Analysis" by John Conway and I am having trouble justifying a certain fact stated in an example:

Let $X$ be the collection of all sequences $\{a_n : n\geq 1\}$ of scalars from $\mathbb{F}$ such that $a_n = 0$ for all but a finite number of values $n$.

In this text we consider $\mathbb{F}$ as either the real numbers or complex numbers, but not any other field. The example then gives the following inner products on $X$.

$$ \langle a_n,b_n\rangle = \sum_{n=1}^\infty a_n\overline{b_n} $$ $$ \langle a_n, b_n\rangle = \sum_{n=1}^\infty \frac1n a_n \overline{b_n} $$ $$ \langle a_n, b_n\rangle = \sum_{n=1}^\infty n^5 a_n \overline{b_n} $$

The example proclaims that these define all the inner products on $X$. This brought two questions to my attention.

  1. How does one find all of the inner products on a vector space?
  2. Why are these the only inner products on $X$ for the example?

First, question 1. There are 6 properties that a function $\langle\cdot, \cdot\rangle:X \times X \rightarrow \mathbb{F}$ has to satisfy in order to be an inner product:

$$ \langle\alpha x + \beta y, z\rangle = \alpha \langle x,z\rangle + \beta \langle y,z\rangle $$ $$ \langle x, \alpha y + \beta z\rangle = \overline{\alpha}\langle x,y\rangle + \overline{\beta}\langle x,z\rangle $$ $$ \langle x,x\rangle \geq 0 $$ $$ \langle x,y\rangle = \overline{\langle y,x\rangle} $$ $$ \langle x,0\rangle = \langle 0,y\rangle = 0 $$ $$ \langle x,x\rangle = 0 \implies x = 0 $$

However, just these properties alone doesn't provide many hints into how to determine the functions. Is there a good strategy for finding the inner products of a vector space?

Secondly, question 2. I don't see why these functions in particular form the collection of all inner products on $X$. At first glance it appears as though any function $f:\mathbb{N} \rightarrow \mathbb{F}$ that is non-zero for all $n\in \mathbb{N}$ would satisfy the conditions for this example. To show this if we let $\langle a_n,b_n\rangle = \sum_{n=1}^\infty f(n)a_n\overline{b_n}$ we have:

$$ \langle\alpha a_n + \beta b_n, c_n\rangle = \sum_{n=1}^{\infty} f(n)(\alpha a_n + \beta b_n)c_n = \alpha \sum_{n=1}^\infty f(n)a_n c_n + \beta \sum_{n=1}^\infty f(n)b_nc_n $$

As required. The second condition is similarly satisfied. The third can be seen since we chose $f(n) \neq 0$ for all $n\in \mathbb{N}$, the fourth and fifth follows easily as well. Lastly the sixth condition is satisfied since we range $a_n$ and $b_n$ over all the natural numbers, and $f(n)$ is non-zero. From this I would have to conclude that there are infinitely many inner products on $X$. What am I missing?

Edit: it turns out I misread the passage the examples above "all define inner products on $X$", not "define all inner products." It makes much more sense now!

  • 1
    $\begingroup$ Those aren't all the possible inner products on $X$. You are right that any $f(n)$ also determines one. But there are even more than that. $\endgroup$ Apr 14 '17 at 14:32
  • 1
    $\begingroup$ You should require $f(n)>0$ for all $n$ if you really want to have inner products. But other than that your construction is correct. $\endgroup$
    – Arnaud D.
    Apr 14 '17 at 14:34

Let $V_k = \{ x:\mathbb N \to \mathbb F \mid \forall n > k : x_n = 0 \}$ and $V = \bigcup_k V_k$. Then, any inner product $\beta$ on $V$ is also an inner product on $V_k$. But inner products on finite dimensional spaces are just positive definite matrices.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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