Show the dual space is a reprensentation of a finite group G.

attempt: Suppose $V^{*} = {\{\phi^*: V → \mathbb{K}}\}$ , where $V^*$ is the set of all linear functions called, functionals. Then the dual space is $V^*$

Then let $\phi:G → GL(V)$ be a linear representation of $ V$. Then let $\phi^* : V → \mathbb{K} \in V^*$. Then we need to show the group action of $G$ on $V$.

$g \dot\ \phi^* : V → \mathbb{K}$ . defined by $g \dot\ \phi^*(v) = \phi^*(g^{-1} \dot\ v)$.

Let $g_1, g_2 \in G, v\in V$ , then

$(g_1g_2 \dot\ \phi^*) = \phi^*((g_1g_2^{-1})v) = \phi^*(g_2^{-1}g_1^{-1} \dot\ v) = \phi^*(g_2^{-1} \dot\ (g_1^{-1} \dot\ v)$ .

And $(g_1(g_2\phi^*)(v) = (g_2\phi^*)(g_1^{-1}v) = \phi(g_2^{-1}g_1^{-1}v)$

I am not sure how I have to prove the dual space is a representation. Or if I am even setting up the problem correctly. Can someone please help me?

  • $\begingroup$ Do you mean "Let $\phi : G\to GL(V)$ be a linear representation of $G$" (not V)? $\endgroup$ – Quinn Greicius Mar 14 '16 at 1:16
  • $\begingroup$ Yes, a typo. Thanks $\endgroup$ – user4294029 Mar 14 '16 at 1:18

If I understand what you are trying to do, you are trying to show that the dual representation of a representation $\rho:G \to \operatorname{GL}(V)$ really is a group homomorphism $\rho^*:G \to \operatorname{GL}(V^*)$. (Defined by $\rho^*_g(\phi^*)(v) = \phi^*(\rho_{g^{-1}}(v))$.) Indeed we have $$\rho^*_{g_1 g_2}(\phi^*)(v) = \phi^*(\rho_{g_2^{-1}g_1^{-1}}(v)) = \phi^*(\rho_{g_2^{-1}}(\rho_{g_1^{-1}}(v))) = \rho^*_{g_1}(\rho^*_{g_2}(\phi^*))(v).$$

  • $\begingroup$ When you use $\phi^*(\rho_{g^{-1}}(v))$ is that equivalent to $\phi^*(\rho(g^{-1}(v))$? Would that complete that it's indeed a representation? We don't have to show it's linear? $\endgroup$ – user4294029 Mar 14 '16 at 1:35
  • $\begingroup$ @user4294029 Yes, I write $\rho_g$ instead of $\rho(g)$ to save parentheses. Also, yes, technically, you should prove that the transpose of a linear map (in this case, the transpose of $\rho_{g^{-1}}$) is a linear map if you've never convinced yourself of this before. $\endgroup$ – Alex Provost Mar 14 '16 at 1:39
  • $\begingroup$ So proving the transpose of a linear map , is a linear map , I would prove $\rho^* : G → GL(V^*)$ is linear? $\endgroup$ – user4294029 Mar 14 '16 at 1:57

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.