I have two questions about representations of compact Lie group.

  1. If all irreducible representations of a compact Lie group are one-dimensinal, then G is abelian.

  2. An infinite compact Lie group has an infinitely number of non-equivalent irreducible representations.

I am very un-familiar with this topic, I know the finite group case, but didn't know how to start these two questions.

  • $\begingroup$ What tools do you have to work with? For 1., consider the adjoint representation when $G$ is non-abelian... $\endgroup$ – Jason DeVito Dec 3 '14 at 19:01
  • $\begingroup$ I don't know adjoint representation... I know representative functions, matrix coefficients, orthogonal relations, Peter and Weyl theorem... I don't know whether they are helpful or not. $\endgroup$ – CYC Dec 3 '14 at 19:13

If you know enough about representations of general compact groups to understand the Peter-Weyl theorem then you're more or less there. Knowing about representations of finite groups is good too; after all, compact groups are particularly nice because you can, more or less, get away with pretending that they're finite.

The trick here to make things simpler is to know a bit of Lie theory. Disclaimer: I'm most definitely not a Lie theorist, so these solutions are probably by no means optimal.

One way to see your first question is to know (or learn!) that every compact Lie group admits a faithful representation. Then identify $G$ with its image in this representation, which is a finite sum of $\mathrm{Aut}(V_i)$, for $\mathrm{Aut}(V_i)$ one-dimensional vector spaces. But then $G$ is just isomorphic to a subgroup of $(\mathbb{C}^\times)^n$ for some $n$ and you're done.

For the second, I believe we can actually say something stronger: a compact Lie group $G$ is finite if and only if it has finitely many irreducible representations (which are necessarily finite-dimensional, as $G$ is compact). You can see this using the Peter-Weyl theorem: the space $L^2(G)$ is isomorphic to the direct sum of the irreducible representations of $G$, with multiplicity. But then $L^2(G)$ is finite-dimensional if and only if $G$ is finite, and we're done.

  • $\begingroup$ Why is $L^2(G)$ isomorphic to direct sum of the irreducible representation of G? and if it is true, why the last statement is true? $\endgroup$ – CYC Dec 3 '14 at 20:24
  • $\begingroup$ $L^2$ under the regular action decomposing as a direct sum of irreducibles with multiplicity is just the Peter-Weyl theorem, which you said you were familiar with. Try thinking about why the second statement is true -- can you think of a canonical way to attach to each element of $G$ a distinct function in $L^2(G)$? Are those functions linearly independent? If you can do that, you're done. $\endgroup$ – PL. Dec 3 '14 at 20:37
  • $\begingroup$ Please give me some details about the second statement, my brain is in a mess right now, I think it could help me understand this topic more. $\endgroup$ – CYC Dec 3 '14 at 20:55

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.