0
$\begingroup$

Assume $G$ is a group scheme over a field $k$ and consider the categories $Rep_G$ of finite dimensional representations of G and $REP_G$ of all representations of G. For two objects $A,B$ in $Rep_G$, is the following formula true?

$$Ext^n_{Rep_G}(A,B)=Ext^n_{REP_G}(A,B)$$

Or equivalently, does every (possibly infinite dimensional) extension

$$0\to B\to X_1\to\dots\to X_n\to A\to 0$$ equivalent to a finite dimensional one?

$\endgroup$
5
  • $\begingroup$ Am I missing something? Doesn't the following work? The exact sequence is equivalent to $0 \to X_1/B \to X_2 \to \ldots\to X_n\to A\to 0$. Per induction all $X_i$ with $i>1$ and $X_1/B$ are finite dimensional. Thus $X_1$ is also finite dimensional. $\endgroup$ Oct 8, 2014 at 13:31
  • $\begingroup$ @JohannesHahn When I use "equivalent", i mean the complex chain should have the two ends fixed...That is, $B$ and $A$ must remain... $\endgroup$
    – KylinChen
    Oct 9, 2014 at 3:13
  • $\begingroup$ Alright maybe. But doesn't my argument show that the sequence is not only "equivalent" to a finite dimensional sequence but is finite dimensional itself. $\endgroup$ Oct 9, 2014 at 14:36
  • $\begingroup$ @JohannesHahn There is no reason for $X_1/B$ to be finite dimensional, is it ? Thus you cannot use induction. $\endgroup$
    – Arkandias
    Oct 9, 2014 at 19:48
  • $\begingroup$ @Arkandias: Oh, you're right. I knew it couldn't be that simple... $\endgroup$ Oct 9, 2014 at 20:01

0

Your Answer

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

Browse other questions tagged or ask your own question.