3
$\begingroup$

$U=span((1,1 ,0),(2 ,0 ,1))$ $W=span((1,1,1),(5,3,1))$

what is $U \cap W$?

can I find the rref of them both and to find the intersections? that mean

$U= \begin{pmatrix} 1 & 0 & 0.5 \\ 0 & 1 & -0.5 \end{pmatrix}$

$W= \begin{pmatrix} 1 & 0 & -1 \\ 0 & 1 & 2 \end{pmatrix}$

so is $U \cap W$=$Span \begin{pmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \end{pmatrix}$

$\endgroup$
2
$\begingroup$

Well, $U \cap W$ is the space consisting of all vectors that can be formed both as a linear combination of the elements of U and V. So, for any $\vec v \in U \cap W$: $$ \vec v = c_1 (1, 1, 0) + c_2 (2, 0, 1) $$ and $$ \vec v = d_1(1, 1, 1) + d_2(5, 3, 1)$$ where $c_1, c_2, d_1, d_2 \in \mathbb{R}$. Does that help?

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ Or maybe it might be easier to use the vectors from the reduced matrix; the computations will be easier. (We have more zeroes.) $\endgroup$ – Martin Sleziak Aug 8 '14 at 5:56
  • $\begingroup$ @Toxicz I have 3 equations with 4 unknowns $\endgroup$ – gbox Sep 17 '14 at 8:33
1
$\begingroup$

From the rrefs you calculated, you know that $\dim U=2$ and $\dim V=2$. (And this can be also seen without calculating rref: We have only two vectors generating a subspace, and neither of these vectors is a multiple of the other one, so they are linearly independent.)

If you make rref from the matrix containing all four vectors, you will find out, that $\dim(U+V)=3$. $$\begin{pmatrix} 1 & 0 & \frac12\\ 0 & 1 & -\frac12\\ 1 & 0 & -1\\ 0 & 1 & 2 \end{pmatrix}\sim \begin{pmatrix} 1 & 0 & 0\\ 0 & 1 & 0\\ 0 & 0 & 1\\ 0 & 0 & 0 \end{pmatrix} $$

So from the formula $\dim(U)+\dim(V)=\dim(U+V)-\dim(U\cap V)$ we can calculate $$\dim(U\cap V)=1.$$ So the intersection is generated by single vector.

If we notice that $$(5,3,1)=5(1,0,\frac12)+3(0,1,-\frac12) = 3(1,1,0)+(2,0,1)$$ then we see that the vector $(5,3,1)$ belongs to both $U$ and $V$, i.e., $(5,3,1)\in U\cap V$.

So now, without any other calculations, we can say that $$U\cap V = \operatorname{span}\{(5,3,1)\}$$


However, in this case we were simply lucky that we only needed one vector and we were able to guess a vector which belongs to both given subspaces.

If we want to avoid guesswork, a reasonable way to go is to solve a system of equations as suggested in Toxicz's answer.

| cite | improve this answer | |
$\endgroup$

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.