2
$\begingroup$

Following a comment of user131781, posted to an answer of this question on MO, I am looking for references to the construction of (co)-free functors from categories into the category of Banach spaces and short maps (which I denote by Ban-1).

... the (originally algebraic) concept of a free object was transferred in a natural way to a plethora of topological versions—-free topological groups, free locally convex spaces ... over a topological space (metric space, uniform space, differentiable manifold, complex manifold) around the middle of the last century. The underlying principle was always the same. You embed your space into a suitable TVS by providing the free vector space with a corresponding structure, then taking the completion which then has a characteristic universal property...

There are countless examples of this ( topological spaces into spaces of measures, metric spaces into free Banach spaces, uniform spaces into uniform measures manifolds into distributions, complex manifolds into analytic functionals,...) and the principle is always the same—-you basically construct an adjoint to a forgetful functor. Of course, you don‘t have to do it or think of it in this way—-most people probably wouldn‘t. However, a small minority is of the opinion that so doing sheds some light on the subject.

Ideally, the source category should itself admit a forgetful functor into Top, the category of topological spaces and continuous functions. However, I'm interested in all such results.

Note:


I would like to point out, that I have never come across a co-free functor with sink as Ban-1. So such a result is also of particular interest to me.

Thanks in advance.

$\endgroup$
6
  • 1
    $\begingroup$ See tandfonline.com/doi/pdf/10.1080/00927878408823035 However, it might be a good exercise for you to prove that the functor $\mathbf{Ban}_1 \rightarrow \mathbf{Set}$ mapping a Banach space to its unit ball has $\ell^1$ as its left adjoint. If I remember rightly, some information on these things can also be found in Borceux's Handbook of Categorical Algebra. $\endgroup$ Commented Oct 2, 2019 at 9:08
  • $\begingroup$ This I have managed to show, but does the forgetful functor $Ban_1\mapsto Top$ have a left-adjoint? $\endgroup$
    – ABIM
    Commented Oct 2, 2019 at 15:11
  • $\begingroup$ This functor does not preserve infinite products, so no. Specifically, the product of $\mathbb{R}$ by itself countably many times is $\ell^\infty$ in $\mathbf{Ban}_1$, but the product in $\mathbf{Top}$ is $\mathbb{R}^{\mathbb{N}}$. $\endgroup$ Commented Oct 2, 2019 at 15:14
  • $\begingroup$ Oh, so some metric structure is indeed needed... Like in the case of the Lipschitz-free space then? $\endgroup$
    – ABIM
    Commented Oct 2, 2019 at 15:21
  • $\begingroup$ Something where the products are compatible. This is also why the forgetful functor taking a Banach space to its underlying set doesn't have a left adjoint. $\endgroup$ Commented Oct 2, 2019 at 15:23

0

You must log in to answer this question.