For a finite group , its irreducible characters form basis for the space of class functions . Define and with .
We introduce on the product. For any and , where (rep. ) is a character of (rep. ), define as
Theorem
The product on is commutative and associative.
\begin{proof}
Define . Then there is a bijection
For associativity, use .
\end{proof}
Furthermore, we can equip the algebra with an inner product
Definition
Define the character map
Remark. For any and any conjugacy class with representative , we can write the character map as
Theorem
The characteristic map is an isometry and an algebra isomorphism. Moreover, and , where is the character of permutation module .
\begin{proof}
A Generalization
Let be a finite group, and let be an associative algebra. If are functions, we may define a bilinear form with values in
If is a class function on and is a class function on , the generalized Frobenius reciprocity still holds, that is
where and with the convention that if .
For any , there is
Also note that .
Since for and , we have
and so is a homomorphism.
Let be the trivial representation of , then
Recall that the permutation module with . Then yields that
To prove that , note that
\eta_{\lambda_1} & \cdots & \eta_{\lambda_1+m-1} \\ \vdots & & \vdots \\ \eta_{\lambda_m-m+1} & \cdots & \eta_{\lambda_m} \end{matrix}\right|$$ and $\mathrm{ch}(\widetilde {\chi^\lambda})=\mathscr s_\lambda$. Now we finish the proof, as $\widetilde {\chi^\lambda}$ is induced $\chi^\lambda$. `\end{proof}` > [!corollary] > > The character table $\chi_{\mathscr p}^\lambda$ is the transition matrix between bases $p_\lambda$ and $\mathscr s_\lambda$ of $\Lambda^n$. ^7ed3d1 `\begin{proof}` It suffices to show $p_{\mathscr p}=\sum_{\lambda}\chi_{\mathscr p}^\lambda \mathscr s_\lambda$. By [[#^492b55|^492b55]] we have\mathscr s_\lambda=\mathrm{ch}(\chi^\lambda)=\sum_{\mathscr p}z_{\mathscr p}^{-1}\chi_{\mathscr p}^\lambda p_{\mathscr p}
and so $\chi_{\mathscr p}^\lambda=\left\langle \mathscr s_\lambda, p_\mathscr p\right\rangle$. It deduces that\left\langle \mathscr s_\lambda,p_\mu\right\rangle =\sum_{\mathscr p} z_\mathscr p^{-1}\chi_{\mathscr p}^{\lambda}\left\langle p_\mathscr p,p_\mu\right\rangle =z_{\mu}^{-1}\chi_{\mu}^\lambda z_{\mu}=\chi_{\mu}^\lambda.
Now we finish the proof. `\end{proof}` > [!corollary] Frobenius character formula > > If $\rho=(\rho_1,\cdots,\rho_m)$ with $\rho_1\geqslant \rho_2\geqslant\cdots\geqslant \rho_m\geqslant 1$, then for any $k\geqslant\ell(\lambda)$ the value $\chi_\rho^\lambda$ is the coefficient of $x_1^{\ell_1}\cdots x_k^{\ell_k}$ in the monomial > > $$ > \prod_{i=1}^m(x_1^{\rho_i}+\cdots+x_k^{\rho_i})\prod_{1\leqslant i< j\leqslant k}(x_i-x_j) > $$ > > where $\ell_i=\lambda_i+k-i$ for $i=1,\cdots,k$. **Remark.** It has been proved in [[10.6 Frobenius Character Formula#^4b468a|^4b468a]]. `\begin{proof}` Recall that $p_{\rho}=\prod_{i=1}^m(x_1^{\rho_i}+\cdots+x_k^{\rho_i})$ and ![[10.8 Symmetric Polynomials#^i6y60w|^i6y60w]] By [[#^7ed3d1|^7ed3d1]], we have $p_{\mathscr p}=\sum_{\lambda}\chi_{\mathscr p}^\lambda \mathscr s_\lambda$ and it deduces that $$\mathrm{LHS}=\sum_{\mu}\chi_{\rho}^\mu\left|\begin{matrix} x_1^{\mu_1+k-1} & ... & x_k^{\mu_1+k-1} \\ \vdots & & \vdots \\ x_1^{\mu_l} & \cdots & x_k^{\mu_k} \end{matrix}\right|.$$ Therefore, the coefficient of $x_1^{\ell_1}\cdots x_k^{\ell_k}$ with $\ell_i=\lambda_i+k-i$ of right hand side is $\chi_{\rho}^\lambda$. `\end{proof}`