Prove Invariance of Intersection Under T

  • Thread starter Thread starter evilpostingmong
  • Start date Start date
  • Tags Tags
    Invariance Proof
evilpostingmong
Messages
338
Reaction score
0

Homework Statement


Suppose T is contained in the set of linear transformations from
V to V. Prove that the intersection of any collection of subspaces of
V invariant under T is invariant under T.



Homework Equations





The Attempt at a Solution


Choose a basis for V. This basis is <v1...vn>.
Two possible subspaces for V are A and B with basis
<v1...vi> for A and <vi-1...vn> for B (2\leqi\leqn)
and we will assume that both are invariant under T.
A basis for the intersection of the bases is <vi-1, vi>.
Because the subspaces A and B are invariant under T, and
vi-1+vi is a linear combination within the span of both invariant subspaces
under T, vi-1+vi is mapped from A to A and from B to B thus it goes
from A and B to A and B, the intersection of both subspaces.
 
Physics news on Phys.org
I think you can give the proof without choosing a basis for V, which I personally find more appealing. To do this, recall that a subset A c V is invariant under T whenever T(A) c A.

Now you want to show that T(A \cap B) \subset A \cap B. You can do this by showing that it is simultaneously a subset of both A and B.
 
Sounds good. Let x\inA\capB
Distributing T, we get T(A\capB)=T(A)\capT(B).
Since x\inA, T(x)\inT(A) and since x\in(B),
T(x)\inT(B) thus T(x)\inT(A)\capT(B)
and since both A and B are invariant under T (thus T(x) x\inA gets mapped to
A and T(x) x\inB gets mapped to B, for T(A) is in A and T(B) is in B for A and B to be invariant), T(x)=cx\inA\capB.
 
Last edited:
evilpostingmong said:
Sounds good. Let x\inA\capB
Good start.

evilpostingmong said:
Distributing T, we get T(A\capB)=T(A)\capT(B).
Probably that's true by linearity, although I don't even think you need that in the proof.
Let's call this statement (*).

evilpostingmong said:
Since x\inA, T(x)\inT(A) and
since x\inB, T(x)\inT(B) thus
T(x)\inT(A)\capT(B)
and since both A and B are invariant under T (thus T(x) x\inA gets mapped to
A and T(x) x\inB gets mapped to B, for T(A) is in A and T(B) is in B for A and B to be invariant),
That part is entirely correct (try finding where you actually needed statement (*), because I couldn't :smile:).

evilpostingmong said:
T(x)=cx\inA\capB.
This may be written down a bit sloppily, with respect to your rigour before. From T(x) being in T(A) you can conclude that it is also in A, and from T(x) being in T(B) it is also in B. So T(x) is in A and B at the same time. QED

Apart from those two minor details, well done!
 
Thanks compuchip, I need to work on writing more elegant proofs.
Sometimes I need to read my own a few times to get them, lol.
 
That's ok, luckily this is a classic example of "practice makes perfect".
So just write a lot of them.

And if you read them and it's hard for you to decipher what you meant, you really need to re-write them :)
 
Prove $$\int\limits_0^{\sqrt2/4}\frac{1}{\sqrt{x-x^2}}\arcsin\sqrt{\frac{(x-1)\left(x-1+x\sqrt{9-16x}\right)}{1-2x}} \, \mathrm dx = \frac{\pi^2}{8}.$$ Let $$I = \int\limits_0^{\sqrt 2 / 4}\frac{1}{\sqrt{x-x^2}}\arcsin\sqrt{\frac{(x-1)\left(x-1+x\sqrt{9-16x}\right)}{1-2x}} \, \mathrm dx. \tag{1}$$ The representation integral of ##\arcsin## is $$\arcsin u = \int\limits_{0}^{1} \frac{\mathrm dt}{\sqrt{1-t^2}}, \qquad 0 \leqslant u \leqslant 1.$$ Plugging identity above into ##(1)## with ##u...
Back
Top