3
$\begingroup$

Given a finite dimensional (real) vector space $V$, and two non-degenerate bilinear forms $(\cdot,\cdot)_1$ and $(\cdot,\cdot)_2$, one can use a basic linear algebra argument to show that there exists a unique invertible linear map $F:V \to V$ such that $$ (F(\cdot),\cdot)_1 = (\cdot,\cdot)_2. $$ Now if we also assume that $V$ carries a representation of a group $G$, with respect to which both $(\cdot,\cdot)_1$ and $(\cdot,\cdot)_2$ are $G$-equivariant, which is to say $$ (v.g,w.g)_1 = (v,w)_1, ~~~~~~ (v.g,w.g)_2 = (v,w)_2, ~~~~~~~ \text{ for all } v,w \in V, g \in G. $$ In this case will the map $F$ always be $G$-equivariant, i.e. will we always have $$ F(v.g) = F(v).g, $$ or do there exist situations where this equality will not hold? Another way to express the question is: Given an equivariant non-degenerate bilinear form $(\cdot,\cdot)$, and an invertible equivariant map $F$, one can produce the new equivariant non-degenerate bilinear form $(F(\cdot),\cdot)$, but can all equivariant non-degenerate bilinear forms be produced in this way from $(\cdot,\cdot)$? In other words can we classify all equivariant non-degenerate bilinear forms in terms of a certain bilinear form $(\cdot,\cdot)$ and all equivariant invertible maps $L$.

$\endgroup$
2
$\begingroup$

This is true: Write $(v,w)_i = \langle A_i(v),\rangle$ where $A_i:V\to V^\ast$ is invertible and $\langle\quad,\quad\rangle:V^\ast\times V \to K$ is the duality pairing. $(\quad,\quad)_i$ is $G$-invariant iff $A_i$ is $G$-equivariant, and $A_1\circ F=A_2$, or $F=A_1^{-1}\circ A_2$ is then $G$-equivariant too.

$\endgroup$

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.