Blog Archive

Wednesday, May 17, 2023

An elegant proof for linear map and its adjoint has conjugate eigenvalue

This article will introduce an elegant way to prove that T and T∗ has a conjugate eigenvalue.

T∗ is adjoint of T, defined by ⟨v,T∗w⟩=⟨Tv,w⟩

The matrix representation of T∗ is the conjugate transpose.

And it has a connection to dual map T′.

We know that the Riesz representation theorem gives the isomorphism V≅V′, V′ is dual space

by ΦV(v)↦⟨−,v⟩

Consider a linear map T:V→W, its dual map T′:W′→V′ defined by T′(φ)=φ∘T

The matrix representation is transpose AT.

And T∗:W→V=ΦV−1T′ΦW

So ΦVT∗=T′ΦW and put a w∈W in

we get ⟨−,T∗(w)⟩=⟨T−,w⟩, and put v∈V in, we get ⟨v,T∗(w)⟩=⟨Tv,w⟩

And you can prove that (T+S)∗=T∗+S∗,(λT)∗=λ―T∗,(TS)∗=S∗T∗

And (T∗)∗=T For T∈L(V), or End(V)

We know that for Ab, the endomorphism can be a ring

And we can view ∗ as a isomorphism from End(V)→Endop(V)

In Endop(V), the S∘opT=T∘S

So (T+S)∗=T∗+S∗,(TS)∗=T∗∘opS∗

show us that adjoint is a ring homomorphism, and the inverse is itself because (T∗)∗=T

So, adjoint is a ring isomorphism, So T−λI is invertible iff (T−λI)∗=T∗−λ―I is invertible,

Thus T and T∗ has a conjugate eigenvalue.

No comments:

Post a Comment

Popular Posts