Let $A$ be an associative algebra over a field $k$ , and $V$ is a finite representation of $A$.

Character $\chi_V(a) = tr|_V\rho(a)$ where $a\in A$

There exists a filtration $0= V_0\subset V_1\subset \cdots \subset V_n =V$ such that $V_i/V_{i-1}$ is irreducible for all nonegative integer $i\leq n$

If there exists another filtration $0= V'_0\subset V'_1\subset \cdots \subset V'_n =V$

Let $W_i = V_i/V_{i-1} , W'_i = V'_i/V'_{i-1}$ , then there exists a permutation $\sigma$ such that $W_i = W'_{\sigma (i)}$ for all nonegative integer $i\leq n$

Here is a proof from the lecture notes introduction to representation theory by Etingof

(Suppose $Char(k)=0$) It's obvious that $\sum_{i=1}^n \chi_{W_i}=\sum_{i=1}^n \chi_{W'_i}$ . And we know that the characters of irreducible representations are linear independent , so the multiplicity of every irreducible representation $W$ of $A$ among $W_i$ and among $W'_i$ are the same. This implies the theorem.

What does the multiplicity of every irreducible representation $W$ of $A$ means?


The number of $W_i$ (or $W_i'$) that are isomorphic to $W$.

Since $\sum_{i=1}^n \chi_{W_i}=\sum_{i=1}^n \chi_{W'_i}$ (because both are equal to the character of $V$), and so, grouping together the terms involving isomorphic $W_i$ (or $W_i'$), $$\sum_Wm_W\chi_W=\sum_{i=1}^n \chi_{W_i}=\sum_{i=1}^n \chi_{W'_i}=\sum_Wm'_W\chi_W,$$ where the first and last sums are over the isomorphism classes of irrreducible representations $W$ and where $m_W$ is the multiplicity of $W$ among the $W_i$ (and $m'_W$ the multiplicity among the $W'_i$), the linear idependence of the $\chi_W$ implies that $m_W=m'_W$ for every $W$.

| cite | improve this answer | |
  • $\begingroup$ Thank you. Why are the multiplicity the same ? Just because the linear independent of the characters ? I am a very beginner , I don't know much about character. Could you please put some details. $\endgroup$ – Syuizen Oct 29 '15 at 16:20
  • $\begingroup$ @Syuizen I've edited my answer with some details. $\endgroup$ – Jeremy Rickard Oct 30 '15 at 11:49
  • $\begingroup$ @JeremyRickard Why are all $\chi_W$ linearly independent? I know that it is true for finite group when the characteristic of $F$ does not divide the carnality of the group. However it seems that the proof for group case is not applicable to associative algebra case. I shall be grateful if you give any hint on this point. Thank you! $\endgroup$ – Hebe Dec 17 '15 at 14:44
  • $\begingroup$ @Hebe There's a proof in the notes of Etingof referred to in the OP, using Jacobson's Density Theorem. $\endgroup$ – Jeremy Rickard Dec 18 '15 at 9:21
  • $\begingroup$ @JeremyRickard Thank you for your comment. I shall look for this note. $\endgroup$ – Hebe Dec 19 '15 at 18:51

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.