2
$\begingroup$

Let $S^1\times\cdots \times S^1( n\text{ times })=\prod_n U(1)\to U(n)$ be the inclusion. This induces a map between classifying spaces

$$ \prod_nBS^1\to BU(n).$$ i.e., $$ (\mathbb{C}P^\infty)^{\times n}\to BU(n).$$

The induced map in cohomology $$ f:H^*(BU(n))=\mathbb{Z}[y_1\cdots,y_n]\to H^*((\mathbb{C}P^\infty)^{\times n})=\otimes_n H^*(\mathbb{C}P^\infty)=\mathbb{Z}[x_1]\otimes \mathbb{Z}[x_2]\otimes\cdots\otimes \mathbb{Z}[x_n], $$ $|x_i|=2$, $|y_i|=2i$ for all $i$.

Is $f$ a monomorphism? Why?

$\endgroup$
  • $\begingroup$ You have to check that all the functors preserve finite limits. I think it's true, but I'm not sure... $\endgroup$ – user40276 Dec 1 '14 at 6:26
1
$\begingroup$

Yes; in fact $f$ is an isomorphism onto $H^{\bullet}(BU(1)^n)^{S_n}$. This is an aspect of the splitting principle for complex vector bundles.

$\endgroup$
  • $\begingroup$ I still do not understand... $\endgroup$ – Shiquan Dec 2 '14 at 1:05
  • 1
    $\begingroup$ @RSQ: I'd like to be able to help you, but just saying "I don't understand" doesn't give me any feedback about how I can improve my explanation. What don't you understand? $\endgroup$ – Qiaochu Yuan Dec 2 '14 at 1:26

Your Answer

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

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