Vector Spaces

←Back to Tech Tree

inventorycoverage

Vector Spaces #

Linear AlgebraDifficulty: ★★☆☆☆Depth: 2Unlocks: 12

Sets with addition and scalar multiplication satisfying axioms.

Interactive Visualization #

⏮◀◀▶▶STEP0.25x1xZOOM

t=0s

Core Concepts #

Key Symbols & Notation #

F (the field of scalars)

Essential Relationships #

Prerequisites (2) #

Vectors Introduction6 atomsSets6 atoms

Unlocks (1) #

Linear Independencelvl 2

Advanced Learning Details

Graph Position #

23

Depth Cost

12

Fan-Out (ROI)

2

Bottleneck Score

2

Chain Length

Cognitive Load #

5

Atomic Elements

35

Total Elements

L2

Percentile Level

L3

Atomic Level

All Concepts (14) #

Teaching Strategy #

Self-serve tutorial - low prerequisites, straightforward concepts.

You already know how to add arrows in the plane and scale them. A vector space is the upgrade that says: “Actually, those rules can apply to lots of things that aren’t arrows”—like polynomials, signals, images, and functions—so long as addition and scaling behave consistently.

TL;DR:

A vector space over a field F is a set V with (1) vector addition making V an abelian group and (2) scalar multiplication by elements of F, linked by distributive/associative axioms. The payoff is that “linear combinations” make sense, unlocking linear independence, bases, and linear maps.

What Is a Vector Space? #

Why we need a definition (not just examples) #

In geometry, vectors look like arrows. But in linear algebra, “vector” really means: an object you can add and scale in a way that behaves like ordinary arithmetic.

That’s powerful because it lets us treat many different domains with one toolkit:

The key is not the shape of objects—it’s the rules.

The official definition #

A vector space is a set VVV together with:

  1. Vector addition: a function +:V×V→V+ : V \times V \to V+:V×V→V

  2. Scalar multiplication: a function ⋅:F×V→V\cdot : F \times V \to V⋅:F×V→V (often written just as avavav)

where FFF is a field (the scalars). Typical choices are R\mathbb{R}R or C\mathbb{C}C. The field matters because we need to add/multiply scalars and have identities/inverses in the scalar world.

We say: “VVV is a vector space over FFF.”

The axioms, grouped by meaning #

You can memorize 8–10 axioms, but it’s better to group them into two clusters plus “compatibility.”

A) Addition behaves like normal addition (abelian group) #

For all u, v, w ∈ V:

  1. 1)Closure under addition: u + v ∈ V
  2. 2)Associativity: (u + v) + w = u + (v + w)
  3. 3)Commutativity: u + v = v + u
  4. 4)Additive identity: there exists 0 ∈ V such that v + 0 = v
  5. 5)Additive inverse: for each v, there exists −v with v + (−v) = 0

This is exactly: “(V, +) is an abelian group.”

B) Scaling is allowed #

For all a∈Fa \in Fa∈F and v ∈ V:

C) Scaling and addition interact consistently (compatibility) #

For all a,b∈Fa,b \in Fa,b∈F and u, v ∈ V:

  1. 1)Distribute scalar over vector addition:

a(u+v)=au+ava(\mathbf{u}+\mathbf{v}) = a\mathbf{u} + a\mathbf{v}a(u+v)=au+av

  1. 2)Distribute field addition over vectors:

(a+b)v=av+bv(a+b)\mathbf{v} = a\mathbf{v} + b\mathbf{v}(a+b)v=av+bv

  1. 3)Associativity of scalar multiplication:

(ab)v=a(bv)(ab)\mathbf{v} = a(b\mathbf{v})(ab)v=a(bv)

  1. 4)Scalar identity acts as identity:

1v=v1\mathbf{v} = \mathbf{v}1v=v

A quick “feel” check #

If you can:

then linear algebra becomes available.

In this lesson, we’ll keep returning to the same habit: test closure + distributivity early—those are frequent failure points, especially in non-geometric examples.

Core Mechanic 1: Addition as an Abelian Group (Closure, Zero, Negatives) #

Why focus on addition first? #

Scalar multiplication is only meaningful if “adding vectors” already forms a stable arithmetic world. The abelian-group requirements guarantee you can:

These are what make expressions like

v1+v2+⋯+vk\mathbf{v}_1 + \mathbf{v}_2 + \cdots + \mathbf{v}_kv1​+v2​+⋯+vk​

well-defined.

Concrete check: does the set contain its own sums? #

Closure under addition is often the first axiom that fails.

Example A (geometric): R2\mathbb{R}^2R2 #

Take u = (1, 2), v = (3, −5). Then

(1, 2) + (3, −5) = (4, −3) ∈ R2\mathbb{R}^2R2.

So closure holds.

Example B (non-geometry): polynomials of degree ≤ 2 #

Let

Then

p(x)+q(x)=(1+x)+(x2−3)=x2+x−2p(x)+q(x) = (1+x) + (x^2-3) = x^2 + x - 2p(x)+q(x)=(1+x)+(x2−3)=x2+x−2

This is still a polynomial of degree ≤ 2. So closure holds.

Example C (common fail): “polynomials of degree exactly 2” #

Let p(x)=x2p(x)=x^2p(x)=x2 and q(x)=−x2q(x)=-x^2q(x)=−x2. Then

p(x)+q(x)=0p(x)+q(x)=0p(x)+q(x)=0

But 0 is not degree exactly 2, so the set is not closed under addition.

The lesson: degree ≤ n works; degree exactly n usually fails.

The zero vector is not always (0,0) #

The additive identity depends on what the objects are.

You should think: “the zero object under addition.”

Additive inverses: subtraction must stay inside #

If v is in the set, then −v must also be in the set.

Polynomial example #

If p(x)=x2+xp(x)=x^2+xp(x)=x2+x, then −p(x)=−(x2+x)=−x2−xp(x)=-(x^2+x)=-x^2-xp(x)=−(x2+x)=−x2−x, which is still a polynomial (and still degree ≤ 2 if we’re in that set).

Function example #

If f(x)=sin⁡xf(x)=\sin xf(x)=sinx, then −f(x)=−sin⁡xf(x)=-\sin xf(x)=−sinx is still a function of the same type.

But beware: if your set is restricted (e.g., “functions that are always nonnegative”), additive inverses typically fail.

Tiny visualization: closure under addition in a function space #

Think of functions as “curves.” Adding functions adds their y-values pointwise.

Take two functions fff and ggg. Their sum is:

(f+g)(x)=f(x)+g(x)(f+g)(x) = f(x) + g(x)(f+g)(x)=f(x)+g(x)

Here’s a simple ASCII snapshot at a few x-values:

xf(x)g(x)(f+g)(x)
0123
1044
2-110

If your set is “all real-valued functions,” the sum is still real-valued at every x, so closure holds.

Interactive canvas idea (guided): Plot two curves (e.g., f(x)=sin⁡xf(x)=\sin xf(x)=sinx, g(x)=0.5xg(x)=0.5xg(x)=0.5x). Add a toggle “show f+g.” Let learners drag a point x₀ and observe the vertical addition of y-values. The closure message becomes: “the result is still a function in the same set.”

Core Mechanic 2: Scalar Multiplication + Compatibility (Distributivity, Identity, Associativity) #

Why scalar multiplication matters #

Scalar multiplication is what lets you talk about size and direction (or more abstractly: intensity, amplitude, coefficient scaling). It’s also what makes linear combinations possible:

a1v1+⋯+akvka_1\mathbf{v}_1 + \cdots + a_k\mathbf{v}_ka1​v1​+⋯+ak​vk​

But scalar multiplication must be consistent with addition; otherwise linear combinations become ambiguous and algebra breaks.

Scalar multiplication closure #

For every a∈Fa \in Fa∈F and v ∈ V, we need av∈Va\mathbf{v} \in Vav∈V.

Non-geometry closure example: polynomials #

Let F=RF=\mathbb{R}F=R and V={polynomials of degree ≤ 2}V={\text{polynomials of degree ≤ 2}}V={polynomials of degree ≤ 2}.

If a=3a=3a=3 and p(x)=x2−1p(x)=x^2-1p(x)=x2−1, then

3p(x)=3(x2−1)=3x2−33p(x)=3(x^2-1)=3x^2-33p(x)=3(x2−1)=3x2−3

Still degree ≤ 2. Closure holds.

Common fail: integer-only scalars with real vectors #

If you try to treat R2\mathbb{R}^2R2 as a vector space over Z\mathbb{Z}Z, you run into a deeper issue: Z\mathbb{Z}Z is not a field (no multiplicative inverses for most integers). Many theorems break, and it’s not a vector space by definition.

So the “F (field)” requirement is not decoration—it’s structural.

Distributivity: the rule that keeps scaling honest #

There are two distributive laws and they fail in many “almost vector spaces.”

1) Scalar over vector sum #

a(u+v)=au+ava(\mathbf{u}+\mathbf{v}) = a\mathbf{u} + a\mathbf{v}a(u+v)=au+av

Function visualization (pointwise):

Let a=2a=2a=2, f(x)=xf(x)=xf(x)=x, g(x)=1g(x)=1g(x)=1.

Same function.

2) Field addition over vectors #

(a+b)v=av+bv(a+b)\mathbf{v} = a\mathbf{v} + b\mathbf{v}(a+b)v=av+bv

Polynomial visualization (coefficient scaling):

Let p(x)=x2+xp(x)=x^2+xp(x)=x2+x and a=2a=2a=2, b=−5b=-5b=−5.

Compute the left side:

(a+b)p=(2−5)p=(−3)p=−3x2−3x(a+b)p = (2-5)p = (-3)p = -3x^2-3x(a+b)p=(2−5)p=(−3)p=−3x2−3x

Compute the right side:

ap+bp=2(x2+x)+(−5)(x2+x)ap + bp = 2(x^2+x) + (-5)(x^2+x)ap+bp=2(x2+x)+(−5)(x2+x)

=(2x2+2x)+(−5x2−5x)= (2x^2+2x) + (-5x^2-5x)=(2x2+2x)+(−5x2−5x)

=−3x2−3x= -3x^2 - 3x=−3x2−3x

They match.

Associativity with field multiplication #

(ab)v=a(bv)(ab)\mathbf{v} = a(b\mathbf{v})(ab)v=a(bv)

This says: it doesn’t matter whether you scale by bbb then aaa, or scale once by ababab.

In Rn\mathbb{R}^nRn this is obvious; in function spaces it is still pointwise obvious:

If (bv)(x)=b⋅v(x)(b\mathbf{v})(x) = b\cdot v(x)(bv)(x)=b⋅v(x), then

a(bv)(x)=a(bv(x))=(ab)v(x)=((ab)v)(x).a(b\mathbf{v})(x) = a(bv(x)) = (ab)v(x) = ((ab)\mathbf{v})(x).a(bv)(x)=a(bv(x))=(ab)v(x)=((ab)v)(x).

Identity scalar #

1v=v1\mathbf{v} = \mathbf{v}1v=v

This is the “do nothing” scale factor.

In polynomials: $1\cdot p(x)=p(x)$.

In matrices: $1\cdot A=A$.

Mini-checklist animation (guided pass/fail) #

When testing “is this a vector space?”, use a consistent order:

  1. What is V? (the set)

  2. What is F? (the scalars)

  3. How are + and scalar multiplication defined?

  4. Closure tests (fast failure):

  1. Zero and inverse (often fails in restricted sets):
  1. Distributivity + identity (consistency)

Interactive canvas idea: Provide toggles for each axiom. For a candidate set (e.g., “nonnegative functions”), clicking “additive inverse” highlights a counterexample: pick f(x)=1f(x)=1f(x)=1 then show −f(x)=−1 not in the set.

This gives learners a reliable mental procedure instead of a memorized list.

Application/Connection: Why Vector Spaces Unlock Linear Independence #

Why vector spaces are the stage for linear algebra #

Most linear algebra concepts are really about linear combinations:

a1v1+⋯+akvka_1\mathbf{v}_1 + \cdots + a_k\mathbf{v}_ka1​v1​+⋯+ak​vk​

To even talk about this expression, you need:

That’s exactly what the vector space axioms guarantee.

Linear combinations are “mixing recipes” #

Think of v₁, v₂ as ingredients and scalars as amounts.

If V is a vector space, then every recipe output is still a valid element of V.

Example: polynomial mixing #

Let V=P2V=P_2V=P2​ (polynomials degree ≤ 2 over R\mathbb{R}R). Take

p1(x)=1p_1(x)=1p1​(x)=1, p2(x)=xp_2(x)=xp2​(x)=x, p3(x)=x2p_3(x)=x^2p3​(x)=x2.

A linear combination is

a⋅1+b⋅x+c⋅x2a\cdot 1 + b\cdot x + c\cdot x^2a⋅1+b⋅x+c⋅x2

which is exactly “an arbitrary quadratic.”

So the vector space structure explains why $

{1, x, x^2}cangenerateallof can generate all of cangenerateallofP_2$.

Connection to linear independence (the next node) #

The next concept—linear independence—asks whether a set of vectors contains redundancy.

A set {v1,…,vk}{\mathbf{v}_1,\dots,\mathbf{v}_k}{v1​,…,vk​} is linearly independent if the only way to get the zero vector from a linear combination is the trivial way.

Formally (over field F):

a1v1+⋯+akvk=0⇒a1=⋯=ak=0.a_1\mathbf{v}_1 + \cdots + a_k\mathbf{v}_k = \mathbf{0} \Rightarrow a_1=\cdots=a_k=0.a1​v1​+⋯+ak​vk​=0⇒a1​=⋯=ak​=0.

This definition relies on:

So: vector spaces are the rules of the game; linear independence is one of the key strategies.

Two non-geometry spaces you’ll see again #

1) Function space (signals) #

Let VVV be the set of continuous functions on [0,1], scalars F=RF=\mathbb{R}F=R.

Vector-space thinking leads directly to Fourier series, least squares, and projections.

2) Matrix space (data) #

Let VVV be the set of all m×nm\times nm×n real matrices, scalars R\mathbb{R}R.

This becomes central in machine learning: datasets, parameter matrices, and gradients live in vector spaces.

A final habit to carry forward #

Whenever you meet a new “vector-like” object (polynomials, functions, sequences, matrices), ask:

That habit will make the next nodes (linear independence, span, basis) feel natural instead of magical.

Worked Examples (3) #

Worked Example 1: Is P₂ (polynomials of degree ≤ 2) a vector space over ℝ? #

Let V = { p(x) : p is a real polynomial with degree ≤ 2 }. Let F = ℝ. Define addition and scalar multiplication in the usual way: (p+q)(x)=p(x)+q(x), and (ap)(x)=a·p(x). Decide whether V is a vector space over F.

  1. Step 1: Identify what must be shown.

    We must verify:

    (1) (V,+) is an abelian group, and

    (2) scalar multiplication by ℝ satisfies closure and compatibility axioms.

  2. Step 2: Check closure under addition.

    Take arbitrary p,q ∈ V.

    Write p(x)=a₂x²+a₁x+a₀ and q(x)=b₂x²+b₁x+b₀.

    Then

    (p+q)(x)=(a₂+b₂)x²+(a₁+b₁)x+(a₀+b₀).

    This is still degree ≤ 2, so p+q ∈ V.

  3. Step 3: Check associativity and commutativity of addition.

    For any polynomials p,q,r, we have pointwise:

    ((p+q)+r)(x)=(p(x)+q(x))+r(x)=p(x)+(q(x)+r(x))=(p+(q+r))(x).

    Similarly p(x)+q(x)=q(x)+p(x) implies p+q=q+p.

    So associativity and commutativity hold.

  4. Step 4: Find the additive identity.

    Let 0(x)=0 for all x (the zero polynomial).

    Then (p+0)(x)=p(x)+0=p(x), so p+0=p.

    Thus the additive identity exists and is in V.

  5. Step 5: Check additive inverses.

    For p(x)=a₂x²+a₁x+a₀, define (−p)(x)=−a₂x²−a₁x−a₀.

    Then (p+(−p))(x)=0 for all x, so p+(−p)=0.

    Also −p still has degree ≤ 2, so −p ∈ V.

  6. Step 6: Check closure under scalar multiplication.

    Take a ∈ ℝ and p(x)=a₂x²+a₁x+a₀.

    Then (ap)(x)=a·a₂x²+a·a₁x+a·a₀, still degree ≤ 2.

    So ap ∈ V.

  7. Step 7: Check distributivity and scalar rules.

    For any a,b ∈ ℝ and p,q ∈ V, pointwise:

    (a(p+q))(x)=a(p(x)+q(x))=ap(x)+aq(x)=((ap)+(aq))(x).

    ((a+b)p)(x)=(a+b)p(x)=ap(x)+bp(x)=((ap)+(bp))(x).

    (ab)p(x)=a(bp(x)) and 1·p(x)=p(x).

    Therefore all compatibility axioms hold.

Insight: P₂ works because “degree ≤ 2” is stable under addition and scaling. Many near-misses fail only because the set isn’t closed (e.g., degree exactly 2, or monic polynomials).

Worked Example 2: Is the set of nonnegative continuous functions a vector space? #

Let V = { f : [0,1] → ℝ | f is continuous and f(x) ≥ 0 for all x }. Let F = ℝ. Operations are pointwise: (f+g)(x)=f(x)+g(x) and (af)(x)=a·f(x). Determine if V is a vector space over ℝ.

  1. Step 1: Try the fastest failure checks (closure + inverses).

    Because V has an inequality restriction (f(x) ≥ 0), additive inverses are suspicious.

  2. Step 2: Check closure under addition.

    Take f,g ∈ V.

    For each x, f(x) ≥ 0 and g(x) ≥ 0, so f(x)+g(x) ≥ 0.

    Also f+g is continuous.

    Thus f+g ∈ V. So addition closure passes.

  3. Step 3: Check closure under scalar multiplication.

    Take a ∈ ℝ and f ∈ V.

    If a ≥ 0, then af(x) ≥ 0, so af ∈ V.

    But the axiom requires closure for all scalars a ∈ ℝ, including negative scalars.

  4. Step 4: Produce a counterexample with a negative scalar.

    Let f(x)=1 (constant function). Then f ∈ V.

    Take a=−1.

    Then (af)(x)=−1 for all x, which is not ≥ 0.

    So af ∉ V.

    Scalar multiplication closure fails.

  5. Step 5: (Optional) Note the additive inverse failure as well.

    If f(x)=1 ∈ V, then −f(x)=−1 is not in V.

    So additive inverses fail too.

Insight: Sets defined by “≥ 0” constraints usually fail to be vector spaces over ℝ because you can’t multiply by negative scalars and stay inside the set.

Worked Example 3: Is ℝ² with a weird addition a vector space? #

Let V = ℝ² and F = ℝ. Define scalar multiplication as usual: a(x,y)=(ax,ay). But define addition by (x₁,y₁) ⊕ (x₂,y₂) = (x₁+x₂+1, y₁+y₂). Is (V, ⊕, ·) a vector space over ℝ?

  1. Step 1: Check closure under ⊕.

    For any (x₁,y₁),(x₂,y₂) ∈ ℝ², we get (x₁+x₂+1, y₁+y₂) ∈ ℝ².

    So closure under addition holds.

  2. Step 2: Find the additive identity with respect to ⊕.

    We need an element e=(e₁,e₂) such that (x,y) ⊕ (e₁,e₂) = (x,y).

    Compute:

    (x,y) ⊕ (e₁,e₂) = (x+e₁+1, y+e₂).

    Set equal to (x,y):

    x+e₁+1=x ⇒ e₁=−1,

    y+e₂=y ⇒ e₂=0.

    So the identity would be e=(−1,0), which exists in V.

  3. Step 3: Check compatibility with scalar multiplication (likely failure).

    A key axiom is distributivity:

    a( u ⊕ v ) should equal au ⊕ av.

    Let u=(0,0), v=(0,0), a=2.

    Compute left side:

    u ⊕ v = (0+0+1,0+0)=(1,0).

    Then a(u ⊕ v) = 2(1,0)=(2,0).

    Compute right side:

    au=(0,0), av=(0,0).

    Then au ⊕ av = (0+0+1,0+0)=(1,0).

    Left ≠ right, since (2,0) ≠ (1,0).

  4. Step 4: Conclude.

    Distributivity fails, so this structure is not a vector space.

Insight: You can invent an “addition” that’s closed and even has an identity, but the moment distributivity fails, linear combinations stop behaving predictably. Distributivity is a core integrity check.

Key Takeaways #

Common Mistakes #

Practice #

easy

Let V be the set of all real 2×2 matrices. With usual matrix addition and scalar multiplication over ℝ, is V a vector space?

Hint: Ask: is the sum of two 2×2 matrices still 2×2? Is scalar multiple still 2×2? Do usual arithmetic properties hold entrywise?

Show solution

Yes. Closure holds because adding/scaling entrywise keeps you in 2×2 matrices. The zero vector is the zero matrix. Additive inverses are negatives of matrices. Distributivity and scalar associativity hold entrywise, inherited from ℝ.

medium

Let V = { (x,y) ∈ ℝ² : x + y = 1 } with usual addition and scalar multiplication over ℝ. Is V a vector space?

Hint: Test closure under addition using two generic points satisfying x+y=1.

Show solution

No. Take u=(1,0) and v=(0,1). Both satisfy x+y=1. But u+v=(1,1), and 1+1=2 ≠ 1, so closure under addition fails.

medium

Let V be the set of all polynomials with real coefficients that satisfy p(0)=0. With usual addition and scalar multiplication over ℝ, is V a vector space?

Hint: Check closure by evaluating (p+q)(0) and (ap)(0). Identify the zero element and additive inverses.

Show solution

Yes. If p(0)=0 and q(0)=0, then (p+q)(0)=p(0)+q(0)=0, so closed under addition. If a∈ℝ, then (ap)(0)=a·p(0)=0, so closed under scalar multiplication. Zero polynomial satisfies p(0)=0 and serves as identity; inverses −p also satisfy (−p)(0)=−p(0)=0. Other axioms follow from usual polynomial arithmetic.

Connections #

Next up: Linear Independence

Related future nodes you’ll likely encounter:

Quality: A (4.4/5)

← back to treebrowse all →