As stated here, Noether's Normalization Theorem states:

Suppose that $R$ is a finitely generated integral domain over a field $K$. Then there exists an algebraically independent subset $\{y_1,y_2,y_3,\dots,y_r\}\in R$ such that $R$ is integral over $K[y_1,y_2,y_3,\dots,y_r]$.

If I correctly understand the statement, $R$ is a $K$-algebra, such that if $ab=0$, then $a=0$ and/or $b=0$. Let its generators be $\{a_1,a_2,\dots,a_n\}$. Then $K[a_1,a_2,\dots,a_n]=R$. $R$ is integral over itself. Hence, $R$ is integral over $K[a_1,a_2,\dots,a_n]$.

  1. Is my reasoning correct?
  2. Does this count as proof of Noether's Normalization Theorem?

Thanks in advance!

  • $\begingroup$ The generators $\{a_1,\ldots,a_n\}$ need not be algebraically independent. $\endgroup$ – Zev Chonoles Jul 22 '13 at 5:24
  • $\begingroup$ @ZevChonoles- sorry this might be a stupid question, but my logic was: if $a_k=$ some combination of the other generators, then every element in $R$ can anyway be constructed by $\{a_1,a_2,\dots,a_n\}\setminus \{a_k\}$. So $\{a_1,a_2,\dots,a_n\}\setminus \{a_k\}\subset \{a_1,a_2,\dots,a_n\}$ can also be the set of generators.. Is it possible to construct an algebraically independent set of generators this way? $\endgroup$ – fierydemon Jul 22 '13 at 5:30
  • $\begingroup$ Ok sorry found a counterexample to my own question. If $\{a,b,c\}$ are the generators, and $a^2=b^3c^4$, then $a$ can't be removed from the list of generators. $\endgroup$ – fierydemon Jul 22 '13 at 5:33

The theorem requires the $y_i$ to be algebraically independent. Your "proof" doesn't even mention this condition. So no, it is not correct.

  • $\begingroup$ @ChrisEagle- Thanks for your answer! $\endgroup$ – fierydemon Jul 22 '13 at 5:34

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.