Is the Intersection of Subspaces Always a Subspace?

In summary, to prove that the intersection of any collection of subspaces of V is a subspace of V, we need to show that there exists an additive identity (0), and that the intersection is closed under addition and scalar multiplication. This can be done by considering an arbitrary collection of subspaces U_i in V, and showing that for any two elements u and w in their intersection, their sum and scalar multiple also belong to the intersection. Therefore, we can conclude that the intersection of any collection of subspaces of V is a subspace of V.
  • #1
tylerc1991
166
0

Homework Statement



Prove that the intersection of any collection of subspaces of V is a subspace of V.

Homework Equations



To show that a set is a subspace of a vector space, I need to show that there exists an additive identity, and that the set is closed under addition and scalar multiplication.

For notation I will let F denote R or C.

The Attempt at a Solution



Consider a collection of subspaces U_1, U_2, ..., U_m of a vector space V.

Clearly any intersection of these subspaces will contain 0, and hence the intersection contains the additive identity.

Now consider two elements, u and w, of any arbitrary intersection of U_1, U_2, ..., U_m. Since u and w are in the intersection of U_1, U_2, ..., U_m, u and w are elements of U_1 and U_2 and ... and U_m. Therefore their addition, u + w must be an element of U_1 and U_2 and ... and U_m. So u + w is an element of any intersection of U_1, U_2, ..., U_m and hence this intersection is closed under addition.

Now consider a scalar c in F and an element u of any arbitrary intersection of U_1, U_2, ..., U_m. Since u is in the intersection of U_1, U_2, ..., U_m, u is an element of U_1 and U_2 and ... and U_m. Therefore cu is an element of U_1 and U_2 and ... and U_m. So cu is an element of any intersection of U_1, U_2, ..., U_m and hence this intersection is closed under scalar multiplication.

Any feedback would be greatly appreciated. If it is sloppy, please let me know what I can do to improve my proof writing. Thanks again!
 
Physics news on Phys.org
  • #2
tylerc1991 said:
I need to show that there exists an additive identity...
I would rephrase that. You need to prove membership, not existence. So just say that you need to show that 0 (the additive identity of V) is a member of each [itex]U_i[/itex].

tylerc1991 said:
Consider a collection of subspaces U_1, U_2, ..., U_m of a vector space V.
That's just a countable collection. You need to consider an arbitrary collection [itex]\{U_i|i\in I\}[/itex], where I might be a "larger" set than the integers.

tylerc1991 said:
Clearly any intersection of these subspaces will contain 0, and hence the intersection contains the additive identity.
Right, but instead of saying "clearly", you can explain how you're using the definitions. You should at least say that the intersection contains 0 because each [itex]U_i[/itex] does.

tylerc1991 said:
Now consider two elements, u and w, of any arbitrary intersection of U_1, U_2, ..., U_m.
An arbitrary intersection could be the intersection of uncountably many subspaces. You seem to have the right idea about how the rest should be done though, but I didn't check every detail, so don't take this as a guarantee that you did everything right. I like to use logical notation for this type of proofs, but if you prefer you can replace logical symbols with words: [tex]x,y\in\bigcap_{i\in I}U_i\ \Rightarrow\ \forall i\in I\quad x,y\in U_i\ \Rightarrow\ \cdots\Rightarrow x+y\in\bigcap_{i\in I}U_i[/tex]

(I'll be away from the computer for at least an hour, and probably the rest of the day, but I'll answer follow-up questions tomorrow in the unlikely event that they haven't already been answered by someone else).
 
  • #3
Fredrik said:
That's just a countable collection. You need to consider an arbitrary collection [itex]\{U_i|i\in I\}[/itex], where I might be a "larger" set than the integers.

I am glad you brought that up, I was curious about this part.

Fredrik said:
I like to use logical notation for this type of proofs, but if you prefer you can replace logical symbols with words

Absolutely, and when I am writing this on paper, I use symbols as well. I just worded everything here because I am not so great with the symbol editor.
 

FAQ: Is the Intersection of Subspaces Always a Subspace?

What is a subspace in linear algebra?

A subspace in linear algebra is a subset of a vector space that satisfies three main properties: it contains the zero vector, it is closed under vector addition, and it is closed under scalar multiplication. This means that any vector in a subspace can be multiplied by a scalar and added to another vector in the subspace, and the result will still be in the subspace. Subspaces are important because they allow us to study smaller structures within larger vector spaces.

How do you determine if a set is a subspace?

To determine if a set is a subspace, you must check if it satisfies the three properties mentioned above. First, check if the zero vector is included in the set. Then, check if any two vectors in the set can be added together and still be in the set. Finally, check if any scalar multiple of a vector in the set is also in the set. If all three properties are satisfied, then the set is a subspace.

Can a subspace have more than one dimension?

Yes, a subspace can have more than one dimension. In fact, the dimension of a subspace is determined by the number of linearly independent vectors it contains. A one-dimensional subspace is a line, a two-dimensional subspace is a plane, and a three-dimensional subspace is a space. However, a subspace can also have a dimension of zero, which is simply the zero vector.

How are subspaces related to matrices?

Subspaces and matrices are closely related in linear algebra. A matrix can be used to represent a linear transformation, which is a function that maps one vector space to another. The columns of a matrix represent the basis vectors for the output vector space. If the columns of a matrix are linearly independent, then the matrix represents a subspace of the output vector space. This allows us to use matrices to study subspaces and understand their properties.

Can a subspace change depending on the basis chosen?

Yes, a subspace can change depending on the basis chosen. This is because the basis vectors determine the span of the subspace. If different basis vectors are chosen, the span of the subspace will also change. However, the subspace will always satisfy the three properties of a subspace, regardless of the basis chosen.

Similar threads

Back
Top