MATHEMATICS S6 UNIT 6: Intersection and Sum of Subspaces.
About Course
In linear algebra, the concepts of Intersection and Sum of Subspaces are fundamental operations that allow us to combine and relate different subspaces within a larger vector space.
Let be a vector space over a field , and let and be two subspaces of .
Intersection of Subspaces (∩W)
Definition: The intersection of two subspaces and , denoted ∩W, is the set of all vectors that are common to both and .
U∩W= {v∈V ∣v∈U and v∈W}
Property: ∩W is always a Subspace.
A key property is that the intersection of any two (or more) subspaces is always a subspace itself. To prove this, we can use the subspace test:
- Contains the zero vector: Since and are subspaces, they both contain the zero vector . Therefore, ∈ U∩W.
- Closed under vector addition: Let , v2∈U∩W. This means ∈U, ∈W, ∈U, and ∈W. Since is a subspace, +v2∈U. Since is a subspace, +v2∈W. Thus, +v2∈U∩W.
- Closed under scalar multiplication: Let ∈U∩W and ∈F (a scalar). This means ∈U and ∈W. Since is a subspace, ∈U. Since is a subspace, ∈W. Thus, ∈U∩W.
Since all three conditions are met, ∩W is a subspace of .
Example: In , let be the x-y plane (= {(x, y,0) ∣x,y∈R}) and be the y-z plane (={(0,y,z) ∣y,z∈R}). Their intersection ∩W would be the y-axis: ∩W= {(0, y,0) ∣y∈R}. This is indeed a subspace of .
Sum of Subspaces (+W)
Definition: The sum of two subspaces and , denoted +W, is the set of all possible sums of a vector from and a vector from .
U+W= {u+w ∣u∈U and w∈W}
Property: +W is always a Subspace The sum of two subspaces is also always a subspace. We can prove this using the subspace test:
- Contains the zero vector: Since and are subspaces, ∈U and ∈W. Therefore, =0V+0V∈U+W.
- Closed under vector addition: Let , v2∈U+W. By definition, =u1+w1 for some ∈U, w1∈W, and =u2+w2 for some ∈U, w2∈W. Then +v2 =(u1+w1) +(u2+w2) = (u1+u2) +(w1+w2). Since is a subspace, +u2∈U. Since is a subspace, +w2∈W. Thus, u1+u2) +(w1+w2) ∈U+W.
- Closed under scalar multiplication: Let ∈U+W and ∈F. By definition, = u+w for some ∈U, w∈W. Then =c(u+w) =cu+cw. Since is a subspace, ∈U. Since is a subspace, ∈W. Thus, +cw ∈U+W.
Since all three conditions are met, +W is a subspace of . The sum +W is often described as the smallest subspace that contains both and . It’s equivalent to the span of the union of bases for and .
Example: In , let be the x-axis (={(x,0,0) ∣x∈R}) and be the y-axis (= {(0, y,0) ∣y∈R}). Their sum +W would be the x-y plane: +W= {(x, y,0) ∣x,y∈R}.
Dimension Formula (Grassmann’s Formula)
For finite-dimensional vector spaces, there’s an important relationship between the dimensions of , , ∩W, and +W:
(U)+dim(W)=dim(U∩W) + dim(U+W)
This formula is extremely useful for finding the dimension of the sum or intersection if the other dimensions are known.
Example using the formula: Using the previous example in :
- : x-y plane, (U)=2
- : y-z plane, (W)=2
- ∩W: y-axis, (U∩W) =1
Using the formula: (U+W) = dim(U)+dim(W)−dim(U∩W) =2+2−1=3. This matches, as +W in this case would be all of , which has dimension 3.
Direct Sum (⊕W)
A special case of the sum of subspaces is the direct sum.
Definition: The sum +W is called a direct sum, denoted ⊕W, if and only if their intersection is the zero vector only: ∩W={0V}
Properties of Direct Sums:
- If =U⊕W, then every vector ∈V can be written uniquely as a sum =u+w, where ∈U and ∈W.
- For a direct sum, the dimension formula simplifies: (U⊕W) = dim(U)+dim(W). This makes sense because there’s no “overlap” (beyond the zero vector) to subtract.
- The concept of direct sum is crucial for decomposing a vector space into simpler components.
Example of Direct Sum: In , let be the x-axis (={(x,0,0) ∣x∈R}) and be the y-z plane (= {(0, y,z) ∣y,z∈R}).
- ∩W= {(0,0,0)}, so their intersection is just the zero vector.
- Therefore, their sum is a direct sum: =U⊕W. Every vector a,b,c) in can be uniquely written as a,0,0)+(0,b,c).
- (U)=1, (W)=2.
- (U⊕W) =1+2=3, which is (R3).
Understanding the intersection and sum of subspaces is crucial for comprehending the structure of vector spaces, linear transformations, and many other advanced topics in linear algebra.
Course Content
Definition.
-
Definitions.
15:45