4
$\begingroup$

Is the following result true? If it is, could you plese give me a reference for it? Thanks in advance!


Let $(G, \mu)$ be any compact abelian group with Haar measure $\mu$ (The case I am interested in is $G=\mathbb{T}^L$ for some countable discrete group $L$), suppose $\phi$ is any measurable symmetric 2-cocycle, i.e., $\phi: G\times G\to\mathbb{T}$ is a measurable map such that $$\phi(x,y)\phi(x+y,z)=\phi(y,z)\phi(x,y+z)$$ holds true for all $(x,y,z)\in E\subset G\times G\times G$ and $$\phi(x,y)=\phi(y,x)$$ holds true for all $(x,y)\in F\subset G\times G$, where $E, F$ are conull sets,

then $\phi$ is a measurable 2-coboundary, i.e., there exists a measurable map $c:G\to \mathbb{T}$ such that $$\phi(x,y)=c(x)c(y)c(x+y)^{-1}$$ holds for all $(x,y)\in S\subset G\times G$, where $S$ is a conull set.

$\endgroup$

1 Answer 1

3
$\begingroup$

One way to see this is to note that $T$ splits from any locally compact abelian group (D.L. Armacost, The Structure of Locally Compact Abelian Groups, 6.16). If the cocycle is commutative, then the associated extension $$ 0 \to T \to E \to G\to 0 $$ is a short exact sequence of abelian groups. Since $T$ splits, the cocycle is a coboundary.

This is discussed in my paper "Locally compact Abelian groups with symplectic self-duality"; see http://www.imsc.res.in/~amri/summaries.html#ssdg in Section 3.

$\endgroup$
4
  • 2
    $\begingroup$ thanks, but I am still worried that the 2-cocycle relation holds almost everywhere, not everywhere in my problem, is it appropriate to have a pure algebraic argument as above to show this? $\endgroup$
    – Jiang
    Commented Aug 27, 2015 at 13:40
  • $\begingroup$ Indeed my answer leaves that issue unaddressed. Isn't there a theorem that says that every measurable cocycle is cohomologous to a continuous one? I can't remember where I have seen such statements. $\endgroup$ Commented Aug 28, 2015 at 3:16
  • 1
    $\begingroup$ thanks, it seems that theorem 10 in ams.org/mathscinet-getitem?mr=414775 is the statement you are talking above? $\endgroup$
    – Jiang
    Commented Aug 28, 2015 at 20:16
  • $\begingroup$ That seems to do the job for second countable compact abelian groups. Thanks for digging it out. $\endgroup$ Commented Aug 29, 2015 at 2:46

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .