8
$\begingroup$

I asked this question on Mathematics Stack Exchange, but got no answer:

Given two vector spaces $V$ and $W$ over a field $K$, what is the dimension of $\operatorname{Hom}_K(V,W)\ $?

To state the partial result I've been able to obtain, let me introduce the notation $$ \alpha:=\dim V,\quad\beta:=\dim W,\quad d(K,\alpha,\beta):=\dim\operatorname{Hom}_K(V,W), $$ $$ \quad\nu:=\operatorname{card}(\mathbb N),\quad\kappa:=\operatorname{card}(K). $$ We can assume $\alpha\ge\nu$.

By the Erdős-Kaplansky Theorem and the inequality $$ \dim\operatorname{Hom}_K(\oplus V_i,\oplus W_j)\le\dim\prod\operatorname{Hom}_K(V_i,W_j), $$ we can also assume $\alpha < \beta$, and we get $$ \kappa^\alpha\beta\le d(K,\alpha,\beta)\le\kappa^\beta $$ (for $\nu\le\alpha < \beta$). Indeed, $\kappa^\alpha\beta$ is the dimension of the space of finite rank linear maps from $V$ to $W$.

For the sake of completeness, let us add explicitly that $$ 1\le\beta\le\alpha\ge\nu $$ implies $$ d(K,\alpha,\beta)=\kappa^\alpha. $$

$\endgroup$

1 Answer 1

9
$\begingroup$

Using $\hom_K(V,W)=W^\alpha$ and

Dimension of infinite product of vector spaces

we get $\dim \hom_K(V,W)=(\#W)^\alpha=((\# K)\cdot\beta)^\alpha$.

$\endgroup$
3
  • $\begingroup$ Dear Fernando: Thanks for you answer. I especially enjoyed the answer mathoverflow.net/a/49572/461 of Todd Trimble's to the question you linked to! $\endgroup$ May 30, 2014 at 9:15
  • $\begingroup$ @Pierre-YvesGaillard you're welcome. I think, however, that my answer has little value, it is just a corollary of Todd's. $\endgroup$ May 30, 2014 at 14:38
  • $\begingroup$ To me, it was invaluable! (Also I now realize - thanks to you, Todd and François - how easy it was to answer my question using the Erdős-Kaplansky Theorem, and how silly my approach was...) $\endgroup$ May 30, 2014 at 14:56

Your Answer

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

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