1
$\begingroup$

Thinking about whether or not there is a natural way to transform $L_\infty$-algebras into $A_\infty$-algebras, I wonder if there is a morphism of cooperads

$\mathcal{A}ss^i\to\mathcal{L}ie^i$

from the Koszul symmetric, associative cooperad $\mathcal{A}ss^i$ to the Koszul Lie cooperad $\mathcal{L}ie^i$. I think there is no such map and my argument is as follows:

Suppose there is such a map. Since $\mathcal{A}ss^i\simeq \mathcal{S}\otimes_H\mathcal{A}ss^*$ and $\mathcal{L}ie^i\simeq \mathcal{S}\otimes_H\mathcal{C}om^*$ (with $\otimes_H$ the tensor product of cooperads), this would imply that there is although a cooperad map

$id_\mathcal{S}\otimes_H \theta : \mathcal{S}\otimes_H\mathcal{A}ss^* \to \mathcal{S}\otimes_H\mathcal{C}om^*$, which in turn would mean that the map

$\theta : \mathcal{A}ss^* \to \mathcal{C}om^*$ exists, too. This however is not true, since both $\mathcal{A}ss$ as well as $\mathcal{C}om$ are finite dimensional in each arity and there is no morphism of operads $\mathcal{C}om\to\mathcal{A}ss$.

Is this a valid argument?

$\endgroup$
5
  • $\begingroup$ Because of the universality of the tensor product. Maybe there are other automorphism on $\mathcal{S}$ then $1$, but that wouldn't change the argument I think. $\endgroup$ Commented Jan 7, 2018 at 4:36
  • 2
    $\begingroup$ It seems nonstandard to call $Lie^i$ the "Koszul Lie cooperad" There is a cooperad governing Lie coalgebras and it's not that one. $\endgroup$ Commented Jan 7, 2018 at 4:38
  • $\begingroup$ I mean the Koszul dual of the Lie operad (in contrast to the linear dual). It's personal taste, I justthink it sounds better. $\endgroup$ Commented Jan 7, 2018 at 4:41
  • 1
    $\begingroup$ Your argument is essentially correct (modulo the 0 morphism from $Com$ to $Ass$). Rather than worrying about automorphisms of $S$ you could just tensor by $S^{-1}$ to get $\theta$. But I'm not sure that this question is mathoverflow appropriate. $\endgroup$ Commented Jan 7, 2018 at 4:42
  • $\begingroup$ It might not be appropriate, right. But I wasn't sure if this is appropriate for math.stackexchange either. I'm fine with a vote for close or a transit to math.stackexchange. Don't want to delete it though, since I think its interesting nontheless. $\endgroup$ Commented Jan 7, 2018 at 4:44

0

You must log in to answer this question.

Browse other questions tagged .