MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Here I am referring to Kasparov's KK-theory, a bivariant functor on the category of separable C* algebras. It is well known that $KK(A, \mathbb{C})$ is K-homology and $KK(\mathbb{C}, B)$ is K-theory, both of which can be calculated for a huge collection of C* algebras (often by topological methods).

I am wondering if anybody knows how to actually calculate $KK(A,B)$ when it is not simply isomorphic to K-theory or K-homology (so I guess I also have to exclude $C_0(\mathbb{R})$. It seems that most of the time the interest in KK groups is not actually calculating them but in constructing specific elements (such as the Dirac / dual Dirac elements in proofs of the Baum-Connes conjecture).

It occurred to me that I have never actually seen anybody explicitly calculate $KK(A, B)$ in nontrivial cases. I am left wondering if this is because nobody knows how to do it or if it just isn't particularly useful to do so. Explanations or references are both appreciated.

share|cite|improve this question
$KK(\mathbb C \oplus \mathbb C,\mathbb C \oplus \mathbb C) = M_2 \mathbb Z$. – Andreas Thom Jan 5 '11 at 16:59
When I attended a course by Jacek Brodzki ( he indeed said that most of the time we'd be interested not in computing KK(A,B) but rather in understanding specific cycles. But I can't say whether he meant it generally, or only for the lecture (back then I though he meant it generally). – Łukasz Grabowski Jan 5 '11 at 23:26
@Andreas: is this more general statement true: $KK(A^m,A^n)=M_{n,m} (KK(A;A))$? – Johannes Ebert Jan 8 '11 at 15:44
up vote 8 down vote accepted

There is Rosenberg-Schochet universal coefficient theorem, which says $KK(A,B)\simeq Ext(K_{\ast}(A),K_{\ast+1}(B))\oplus Hom(K_{\ast}(A),K_{\ast}(B))$ (not canonically) when $A$ is $KK$-equivalent to a commutative algebra. It was proved in The Künneth theorem and the universal coefficient theorem for Kasparov’s generalized K-functor, Duke Math. J., 55(1987), 431–474.

share|cite|improve this answer
* gets interpreted as code for italics, so you should use \ast on MO instead. Hopefully I formatted that correctly. – Qiaochu Yuan Jan 5 '11 at 16:33
This is indeed the sort of thing I had in mind - thanks! – Paul Siegel Jan 12 '11 at 13:02

Hi Paul, I do not know if this is interesting (enough) for you, but in 'The operator K-functor and extensions of C*-algebras' Kasparov shows that for suitable A and B (I guess A,B seperable, A nuclear) there exists an isomorphism from $\mathrm{KK}(A,B)$ onto a group $\mathrm{Ext}(A,C_0(\mathbb{R})\otimes B)$ of 'stable equivalence classes of extensions' $$0\xrightarrow{}{} C_0(\mathbb{R})\otimes B\otimes\mathcal{K}\xrightarrow{}{} C \xrightarrow{}{} A \xrightarrow{}{} 0\text{ .}$$. Maybe the RHS is not concrete enough for this to qualify as an answer.

share|cite|improve this answer
Hi Dan! This is indeed a useful answer as well, though as you feared I don't really know how to concretely describe extension groups for C* algebras, either. If you know any good tools for computing extension groups for the sorts of C* algebras which make your isomorphism work, that would interest me as well. – Paul Siegel Jan 12 '11 at 13:04

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.