4
$\begingroup$

Let $\cal K$ be a 2-category, supporting two Yoneda structures, induced by a pseudoadjunction $P^\sharp\dashv P : {\cal K}^\text{coop}\to \cal K$; this means that there is such an adjunction, or in other words $$ {\cal K}(A, PB)\cong {\cal K}(B,P^\sharp A)^\text{op} \tag{$\star$} $$ naturally in all arguments, and that

  • We are given an ideal of $P$-admissible 1-cells and an ideal of $P^\sharp$-admissible 1-cells, and "Yoneda embedding maps" $y_A : A\to PA$ and $y^\sharp_A : A\to P^\sharp A$.
  • For every admissible $f : A\to B$ with admissible domain, we are given arrows $B(f,1)$ exhibiting the left extension of $y_A$ along $f$, and $B(1,f)$ exhibiting the right extension of $y_A^\sharp$ along $f$.

These data satisfy axioms YS1, YS2, YS3 that you can read on the $n$Lab page.

I'm trying to prove two things:

  1. $P$-admissible objects and $P^\sharp$-admissible objects coincide;
  2. If $f : A \leftrightarrows B : u$ is an adjunction, $B(f,1) : B\to PA$ and $B(1,u) : A\to P^\sharp B$ correspond each other under the adjunction $(\star)$.

In $\bf Cat$, with $P=[\,\_°, Set]$ and $P^\sharp=[\,\_\,, Set]°$, 1. and 2. are true.

$\endgroup$
  • $\begingroup$ Isn't there already some counter example to 1 in the example of Cat ? One admissibility conditions means that hom (x,F y) is small for all x,y the other that hom (F x, y) is small for all x,y. $\endgroup$ – Simon Henry Jun 13 '18 at 9:42
  • $\begingroup$ Hm, what I meant is that admissible objects are the same for both $P, P^\sharp$, which seems true. $\endgroup$ – Fosco Loregian Jun 13 '18 at 11:07

Your Answer

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

Browse other questions tagged or ask your own question.