•This lesson teaches three core ideas in linear algebra: linear combinations, span, and basis vectors. A linear combination is when you multiply vectors by numbers (scalars) and add them. The span is the set of all places you can reach using those linear combinations. A basis is a special set of vectors that both spans the space and doesn’t contain any vector that can be made from the others.
•In 2D, the standard basis vectors are i-hat and j-hat, written as $(\begin{pmatrix}1\\0\end{pmatrix}, \begin{pmatrix}0\\1\end{pmatrix})$. Any 2D vector can be built by scaling and adding these two. For example, $2\hat{i} + 3\hat{j}$ equals $\begin{pmatrix}2\\3\end{pmatrix}$, and this reaches the point $(2,3)$. These two vectors are enough to cover the whole 2D plane.
•A linear combination looks like $a\mathbf{v} + b\mathbf{w}$, where $a$ and $b$ are numbers and $\mathbf{v}, \mathbf{w}$ are vectors. For example, if $\mathbf{v}=\begin{pmatrix}1\\0\end{pmatrix}$ and $\mathbf{w}=\begin{pmatrix}0\\1\end{pmatrix}$, then with $a=2, b=-1$, we get $2\mathbf{v} + (-1)\mathbf{w} = \begin{pmatrix}2\\-1\end{pmatrix}$. By letting $a$ and $b$ be any real numbers, you can reach every point in the plane. This is why the span of $\hat{i}$ and $\hat{j}$ is the entire 2D plane.
•If two vectors lie on the same line through the origin, their span is only that line. For instance, $\begin{pmatrix}2\\2\end{pmatrix}$ and $\begin{pmatrix}1\\1\end{pmatrix}$ point in the same direction, so any combination stays on that line. Using $a=3, b=-1$ gives $3\begin{pmatrix}1\\1\end{pmatrix} + (-1)\begin{pmatrix}2\\2\end{pmatrix} = \begin{pmatrix}1\\1\end{pmatrix}$. You never leave that line no matter which numbers you pick.
•In 3D, the span of two non-parallel vectors is a plane through the origin. For example, $\begin{pmatrix}1\\0\\0\end{pmatrix}$ and $\begin{pmatrix}0\\1\\0\end{pmatrix}$ span the $xy$-plane: $a\begin{pmatrix}1\\0\\0\end{pmatrix} + b\begin{pmatrix}0\\1\\0\end{pmatrix} = \begin{pmatrix}a\\b\\0\end{pmatrix}$; picking $a=2,b=3$ gives $\begin{pmatrix}2\\3\\0\end{pmatrix}$. Three well-chosen vectors usually span all of 3D. But if one is a combination of the others, you only get a plane.
Why This Lecture Matters
Understanding linear combinations, span, and basis vectors is essential because they are the language of linear algebra. Anyone studying data science, computer graphics, robotics, physics, or engineering will meet these ideas daily. They solve practical problems such as expressing positions in different coordinate systems, checking if a set of directions covers all necessary motions, and ensuring there is no redundancy in models. For example, in 3D graphics, the standard basis describes x, y, z directions, and any point or movement is a linear combination of these basis vectors; verifying independence makes sure your coordinate system is stable and non-degenerate. In robotics, joints create movements that can be treated as vectors; the span tells whether a robot’s arm can reach certain positions, and the basis describes the minimal control directions needed.
In data science and machine learning, features are like axes in a high-dimensional space. Linear combinations explain models like linear regression, where predictions are combinations of features; the span of selected features decides what patterns can be represented. Basis and dimension help with dimensionality reduction and understanding rank in matrices, which affects solution uniqueness and stability. Mastering these concepts builds the foundation for matrices and linear transformations next, which are the engines behind rotations, scalings, projections, and many learning algorithms. Knowing this not only improves problem solving but also strengthens your career readiness in any field that models the world with vectors and linear rules.
Lecture Summary
Tap terms for definitions
01Overview
This lesson builds a clear, picture-first understanding of three central ideas in linear algebra: linear combinations, span, and basisvectors. You start with simple 2D arrows—vectors like i-hat and j-hat, written as (1,0) and (0,1)—and see how multiplying them by numbers and adding them together can land you at any point in the plane. Formally, any expression of the form a⋅v + b⋅w, where a and b are numbers and v and w are vectors, is called a linear combination; for example, 2⋅(1,0) + 3⋅(0,1) = (2,3) shows how these rules move you to a specific point. By letting a and b range over all real numbers, the set of all such results fills an entire region, which is called the span of the vectors; for instance, the span of (1,0) and (0,1) is the whole 2D plane, while the span of (1,1) and (2,2) is just the line they lie on.
The lesson then shifts to 3D to deepen intuition. In 3D, the span of two non-parallel vectors is a plane through the origin, and the span of three well-chosen vectors is the entire 3D space. But if one of those three vectors can be written as a combination of the other two, the span shrinks down to a plane, not the full space. This idea—whether some vector can be created from others—is captured by linear independence: a set of vectors is linearly independent if none of them can be formed by a linear combination of the others. For example, (1,0) and (0,1) are independent in 2D; there’s no way to scale and add one to get the other unless you use zero scaling.
Using these ideas, the lesson defines a basis. A basis for a space is a set of vectors that spans the space and is linearly independent. In 2D, many pairs of vectors can be a basis; for example, (3,−2) and (1,2) are not multiples of each other and can be scaled and added to reach any point in the plane. However, any three vectors in 2D cannot all be independent, so a triple like (3,−2), (1,2), and (2,2) is not a basis because one of them is a combination of the other two. The smallest number of basisvectors needed to describe a space is its dimension, so 2D needs two and 3D needs three.
The audience for this lesson is beginners who want a visual and intuitive feel for linear algebra. You should be comfortable with basic arithmetic and simple equations. Knowing what a vector is—as a directed arrow or as a pair or triple of numbers—helps, but all needed ideas are introduced from scratch. No coding or advanced math is required.
After completing this lesson, you will be able to: (1) write and recognize linear combinations, like a⋅v + b⋅w, and compute specific results; (2) describe and visualize spans as lines, planes, or full spaces depending on how vectors relate; (3) test whether vectors are linearly independent by checking whether one can be made from the others; (4) identify or construct bases for 2D and 3D spaces; and (5) explain why the dimension equals the size of a minimal basis. These skills apply to many real problems, such as expressing coordinates, understanding directions in space, and laying the groundwork for matrices and transformations.
The structure flows from basic building blocks to bigger structures. First, it defines linear combinations using the standard basisvectors in 2D as a clear, concrete example. Next, it builds the idea of span by showing how varying the numbers a and b paints a line, a plane, or all of space. Then it introduces linear independence and uses it to define a basis—an independent, spanning set. Finally, it explains dimension as the count of vectors in a minimal basis, tying everything back to the simple visual of how many independent directions you truly need to reach every point.
Key Takeaways
✓Always start with the question: what can my chosen vectors reach? Compute linear combinations like $a\mathbf{v}+b\mathbf{w}$ and look at the shape they fill; for example, try $a=2,b=-1$ with $\begin{pmatrix}1\\0\end{pmatrix},\begin{pmatrix}0\\1\end{pmatrix}$ to get $\begin{pmatrix}2\\-1\end{pmatrix}$. If they are multiples, expect only a line. If not, in 2D you get the plane; in 3D, two give a plane and three can give all space.
✓To test if a target vector is in a span, set up equations by matching coordinates. For example, for $\begin{pmatrix}4\\0\end{pmatrix}$ as $a\begin{pmatrix}3\\-2\end{pmatrix} + b\begin{pmatrix}1\\2\end{pmatrix}$, solve $3a+b=4$ and $-2a+2b=0$ to get $a=1,b=1$. A solution means it’s inside; no solution means it’s outside. This turns geometry into simple algebra.
✓Check independence to avoid redundancy. Solve $a_1\mathbf{v}_1 + \cdots + a_k\mathbf{v}_k = \mathbf{0}$ and see if only the all-zero solution works; for instance, with $\begin{pmatrix}1\\0\end{pmatrix}$ and $\begin{pmatrix}0\\1\end{pmatrix}$, only $a=b=0$ gives $\begin{pmatrix}0\\0\end{pmatrix}$. If nonzero coefficients work, the set is dependent. Remove redundant vectors to find a basis.
✓Use the standard basis to build intuition. In 2D, write any vector as $x\hat{i}+y\hat{j}$; e.g., $x=2,y=3$ gives $\begin{pmatrix}2\\3\end{pmatrix}$. In 3D, use $x\begin{pmatrix}1\\0\\0\end{pmatrix} + y\begin{pmatrix}0\\1\\0\end{pmatrix} + z\begin{pmatrix}0\\0\\1\end{pmatrix} = \begin{pmatrix}x\\y\\z\end{pmatrix}$. This makes coordinates concrete and visual.
✓Remember: two non-collinear vectors span 2D, but not 3D. In 3D, two vectors form a plane: $a\mathbf{v}+b\mathbf{w}$. Add a third independent vector $c\mathbf{u}$ to reach anywhere: $a\mathbf{v}+b\mathbf{w}+c\mathbf{u}$; for example, $2,3,-1$ on the standard basis gives $\begin{pmatrix}2\\3\\-1\end{pmatrix}$. This helps you plan how many directions you need.
Glossary
Vector
A vector is like an arrow with direction and length. In 2D, it is written as two numbers, and in 3D, three numbers. These numbers tell how far to move along each axis. Vectors always start at the origin in math drawings unless stated otherwise. You can stretch them (scale) and add them together.
Scalar
A scalar is just a plain number used to stretch or shrink a vector. Positive scalars keep the direction, negative ones flip it. Zero makes the zero vector. Scalars tell you how many copies of a vector to take. They are the multipliers in a linear combination.
Linear combination
This is when you multiply vectors by scalars and add them. It’s the basic recipe for making new vectors from known ones. The general form is $a\mathbf{v} + b\mathbf{w}$. Changing $a$ and $b$ changes the result. It connects algebra to geometry by showing reachable points.
Span
Span is the set of all vectors you can reach using linear combinations of some starting vectors. It shows the full area, line, plane, or space you can cover. If vectors point in only one direction, the span is a line. If they point in different directions, the span can be a plane or all of space. Span always includes the origin.
Version: 1
•Linear independence means no vector in the set can be made by combining the others. For example, $\begin{pmatrix}1\\0\end{pmatrix}$ and $\begin{pmatrix}0\\1\end{pmatrix}$ are independent, because there are no numbers $a,b$ (except $a=b=0$) that make $a\begin{pmatrix}1\\0\end{pmatrix} + b\begin{pmatrix}0\\1\end{pmatrix} = \begin{pmatrix}0\\0\end{pmatrix}$; for a check, $a=2,b=-2$ gives $\begin{pmatrix}2\\-2\end{pmatrix}\neq\begin{pmatrix}0\\0\end{pmatrix}$. Independence matters because it avoids redundancy. It ensures each basis vector brings a new direction.
•A basis is an independent set of vectors that spans the space. In 2D, any two non-collinear vectors can be a basis, such as $\begin{pmatrix}3\\-2\end{pmatrix}$ and $\begin{pmatrix}1\\2\end{pmatrix}$. For instance, $1\cdot\begin{pmatrix}3\\-2\end{pmatrix} + 1\cdot\begin{pmatrix}1\\2\end{pmatrix} = \begin{pmatrix}4\\0\end{pmatrix}$ shows how to reach a point like $(4,0)$. These two cover the plane and are not multiples of each other.
•If you add a third vector in 2D, the set cannot be independent. For example, $\begin{pmatrix}3\\-2\end{pmatrix}$, $\begin{pmatrix}1\\2\end{pmatrix}$, and $\begin{pmatrix}2\\2\end{pmatrix}$ live in 2D, so one must be a combination of the others. Indeed, $\begin{pmatrix}2\\2\end{pmatrix} = \tfrac{1}{2}\begin{pmatrix}3\\-2\end{pmatrix} + \tfrac{1}{2}\begin{pmatrix}1\\2\end{pmatrix}$; checking numbers, $\tfrac{1}{2}\cdot\begin{pmatrix}3\\-2\end{pmatrix} + \tfrac{1}{2}\cdot\begin{pmatrix}1\\2\end{pmatrix} = \begin{pmatrix}2\\0\end{pmatrix}$ is not correct, so adjust to $\begin{pmatrix}2\\2\end{pmatrix} = 1\cdot\begin{pmatrix}1\\2\end{pmatrix} + 1\cdot\begin{pmatrix}1\\0\end{pmatrix}$ where $\begin{pmatrix}1\\0\end{pmatrix}$ equals $\tfrac{2}{7}\begin{pmatrix}3\\-2\end{pmatrix} + \tfrac{1}{7}\begin{pmatrix}1\\2\end{pmatrix}$ to illustrate dependency numerically. This shows redundancy appears with too many vectors in a low-dimensional space.
•Dimension counts the smallest number of basis vectors needed to cover the space. The plane is 2D because two independent vectors suffice. 3D space needs three independent vectors, such as $\begin{pmatrix}1\\0\\0\end{pmatrix}$, $\begin{pmatrix}0\\1\\0\end{pmatrix}$, and $\begin{pmatrix}0\\0\\1\end{pmatrix}$. For example, $2\begin{pmatrix}1\\0\\0\end{pmatrix} + 3\begin{pmatrix}0\\1\\0\end{pmatrix} + (-1)\begin{pmatrix}0\\0\\1\end{pmatrix} = \begin{pmatrix}2\\3\\-1\end{pmatrix}$ reaches $(2,3,-1)$.
•Geometrically, a span is like drawing all paths you can make with certain arrows from the origin. Two independent arrows in 2D give you the full floor (the whole plane). Two arrows on the same line only let you walk along that line. In 3D, two independent arrows give a sheet (a plane), and three well-chosen arrows give the whole room.
•Algebraically, to test if a vector is in a span, you solve for the scalars. If you want $\mathbf{u}$ to equal $a\mathbf{v} + b\mathbf{w}$, you solve the system of equations from matching coordinates; for example, if $\mathbf{u}=\begin{pmatrix}4\\0\end{pmatrix}$, $\mathbf{v}=\begin{pmatrix}3\\-2\end{pmatrix}$, $\mathbf{w}=\begin{pmatrix}1\\2\end{pmatrix}$, then you solve $a\cdot3 + b\cdot1 = 4$ and $a\cdot(-2) + b\cdot2 = 0$, which gives $a=1, b=1$, and indeed $1\cdot\begin{pmatrix}3\\-2\end{pmatrix} + 1\cdot\begin{pmatrix}1\\2\end{pmatrix} = \begin{pmatrix}4\\0\end{pmatrix}$. A solution means it’s in the span; no solution means it’s not. This connects geometry to solving equations.
•Linear independence can be checked by asking if the only solution to $a\mathbf{v} + b\mathbf{w} = \mathbf{0}$ is $a=b=0$. For instance, with $\mathbf{v}=\begin{pmatrix}1\\0\end{pmatrix}$ and $\mathbf{w}=\begin{pmatrix}0\\1\end{pmatrix}$, the equations are $a=0$ and $b=0$ to get $\begin{pmatrix}0\\0\end{pmatrix}$; checking nonzero values like $a=2, b=-2$ yields $\begin{pmatrix}2\\-2\end{pmatrix}\neq\begin{pmatrix}0\\0\end{pmatrix}$. If a nonzero solution exists, the vectors are dependent. This prevents choosing duplicate directions.
•The standard basis gives a simple coordinate system. In 2D, $\hat{i}=\begin{pmatrix}1\\0\end{pmatrix}$ and $\hat{j}=\begin{pmatrix}0\\1\end{pmatrix}$ let any vector $\begin{pmatrix}x\\y\end{pmatrix}$ be written as $x\hat{i} + y\hat{j}$; with $x=2, y=-1$, you get $2\hat{i} + (-1)\hat{j} = \begin{pmatrix}2\\-1\end{pmatrix}$. The numbers $x$ and $y$ are how far you move along each basis arrow. This is how coordinates work.
•The main message: linear combinations are the building blocks, span is the playground they cover, and a basis is the smallest, non-redundant set of building blocks. Without these ideas, coordinates and dimensions would be confusing. With them, you can describe positions and directions clearly in any dimension. This sets you up for matrices next, which are built from how they move basis vectors.
02Key Concepts
01
Linear Combination: A linear combination is when you scale vectors and then add them together to get a new vector. You might write it as av+bw, which means multiply v by a and w by b, then add; for example, with v=(10), w=(01), a=2, and b=3, we get 2v+3w=(23). It’s like mixing paint colors: each vector is a color, and the numbers say how much of each to pour. This is the basic way we build new directions and positions from known ones. Without linear combinations, we couldn’t describe how to reach new points based on simple building blocks.
02
Standard Basis in 2D: The standard basisvectors are i^=(10) and j^=(01); for example, 2i^+(−1)j^=(2−1) shows how coordinates work. They point along the x- and y-axes, making them easy to visualize. Any vector(xy) can be written as xi^+yj^; for instance, x=3,y=4 gives 3i^+4j^=(34). This pair spans the entire 2D plane and is independent, so it’s a basis. It’s the simplest coordinate system for the plane.
03
Span: The span of a set of vectors is the collection of all linear combinations you can make from them. If you have v and w, then span{v,w}={av+bw:a,b∈R}; for example, with v=(10), w=(01), the span is the whole plane, and a=2,b=−1 gives (2−1). Geometrically, span is the set of places you can reach using the given arrows. If the vectors lie on one line, the span is just that line; if they point in different directions, the span can be a plane or even space. Span tells you how much territory your vectors cover.
04
Collinearity in 2D: If two vectors are on the same line through the origin, any combination stays on that line. For example, (22) is 2⋅(11), so both are collinear; choosing a=3,b=−1 gives 3(11)+(−1)(22)=(11), still on the same line. In that case, the span is not the whole plane, just that single line. You need vectors that point in different directions to fill the plane. This shows why independence matters.
05
Span in 3D: In three dimensions, two non-parallel vectorsspan a plane through the origin. For example, with a=100 and b=010, combinations look like sa+tb=st0; with s=2,t=3 we get 230, which lies in the xy-plane. Three well-chosen vectors can cover all of 3D. But if one is a combination of the others, the span collapses to a plane. The span shows the dimensional shape your vectors create.
06
Linear Independence: A set of vectors is linearly independent if none of them can be formed from a combination of the others. Algebraically, the only solution to a1v1+⋯+akvk=0 must be a1=⋯=ak=0; for example, with v1=(10), v2=(01), the equations force a1=a2=0 to get (00). If a nonzero combination gives the zero vector, the set is dependent. Independence ensures there’s no redundancy. It’s crucial for building a basis.
07
Basis: A basis is a set of vectors that is both spanning and independent. In 2D, (3−2) and (12) form a basis because they point in different directions and can reach any point, for example 1⋅(3−2)+1⋅(12)=(40). In 3D, three independent vectors make a basis for space. A basis is like a minimal set of direction arrows that can build everything in the space. It makes coordinates possible and unique.
08
Redundancy in 2D Triples: In 2D, any three vectors must be linearly dependent. For instance, with (3−2), (12), and (22), one can be made from the others; numerically, you can solve α(3−2)+β(12)=(22) to find α=72, β=712, giving 72(3−2)+712(12)=(22). This shows the third adds no new direction. Therefore, such a triple cannot be a basis. Removing redundancy restores independence.
09
Dimension: The dimension of a space is the size of a smallest basis for that space. The plane is 2D because two independent vectors are enough, while space is 3D because three are needed. For example, (10),(01) form a 2D basis, and 100,010,001 form a 3D basis; combining 2100+3010+(−1)001=23−1 reaches (2,3,−1). Dimension counts how many independent directions you need. It’s a fundamental summary of a space’s size.
10
Geometric Picture of Span: Think of vectors as arrows from the origin. The span is the entire shape you can sweep out by stretching and adding those arrows, like painting with two brushes. With two non-collinear 2D arrows, you cover the whole floor; for instance, a(10)+b(01) gives (ab), and a=2,b=−1 gives (2−1). With collinear arrows, you only trace a line. Geometry makes the algebra feel natural.
11
Testing Membership in a Span: To check if a vectoru is in span{v,w}, solve av+bw=u for a,b. For example, if v=(3−2), w=(12), and u=(40), solving 3a+1b=4 and −2a+2b=0 gives a=1,b=1, confirming membership. If a solution exists, u lies in the span; otherwise, it doesn’t. This makes the idea computable. It links vectors to simple systems of equations.
12
Independence vs. Dependence: Independence means no unnecessary vectors; dependence means at least one is a mix of the others. In 3D, three independent vectors form a basis, but if the third equals pv+qw, the set is dependent; e.g., if u=110, v=100, w=010, then u=1⋅v+1⋅w shows dependence with p=q=1. Dependence shrinks your span. Independence keeps it as large as possible for the given count. This is why basisvectors must be independent.
13
Uniqueness of Coordinates in a Basis: In a basis, each vector in the space has exactly one representation as a combination of the basisvectors. For instance, in 2D with basis{(3−2),(12)}, the point (40) equals 1⋅(3−2)+1⋅(12) and cannot be written with different coefficients, which you can check by solving and seeing a=1,b=1 is unique. If there were two different ways, subtracting would yield a nonzero combination equaling zero, contradicting independence. This uniqueness makes coordinates well-defined. It’s a key reason bases are powerful.
14
Planes Through the Origin: Any span of two independent 3D vectors makes a plane that passes through the origin. The formula av+bw always gives points that start from the origin because vectors begin at the origin; for example, 2100+3010=230 sits in the xy-plane through the origin. Adding a third independent vector lets you lift off that plane to fill space. If the third is dependent, you stay stuck in the plane. This origin-anchored structure is a hallmark of spans and subspaces.
15
Dimension Limits Independence: In an n-dimensional space, any set with more than nvectors is dependent. In 2D, any 3 vectors must be dependent; you can solve to express one as a combination of the others, e.g., (22)=72(3−2)+712(12) as a concrete instance. In 3D, any 4 vectors are dependent. This counting idea is a practical shortcut. It helps you predict when redundancy must appear.
03Technical Details
Foundational structure: linear combinations, span, independence, basis, and dimension
Linear combinations
🎯 One-line definition: A linear combination is an expression where you multiply vectors by numbers and add the results.
🏠 Everyday analogy: It’s like making a smoothie from fruits. Each fruit type is a vector, and how many scoops you add are the numbers (scalars). Mixing different amounts gives a new flavor (a new vector).
🔧 Technical explanation: Given vectors v and w, any vector of the form av+bw is called their linear combination; for instance, with v=(10), w=(01), a=2, b=−1, we get 2v+(−1)w=(2−1), a concrete point.
💡 Why it matters: Linear combinations are the basic actions that let us build all other vectors from a small set. Without them, we couldn’t describe where we can go starting from known directions.
📝 Concrete example: In 2D, 2i^+3j^ equals (23) using i^=(10) and j^=(01), so the numbers 2 and 3 say “go 2 units in x, 3 units in y.”
Properties of combining:
Scalar multiplication stretches or flips a vector: c(xy)=(cxcy); for example, with c=−2 and (xy)=(3−1), you get −2(3−1)=(−62).
Vector addition combines coordinates: (x1y1)+(x2y2)=(x1+x2y1+y2); for instance, (2−3)+(−15)=(12).
Together they give linear combinations: a(x1y1)+b(x2y2)=(ax1+bx2ay1+by2); for example, with a=2,b=3,(x1y1)=(14),(x2y2)=(−21), you get (2⋅1+3⋅(−2)2⋅4+3⋅1)=(−411).
Span
🎯 One-line definition: The span of vectors is the set of all linear combinations you can make from them.
🏠 Everyday analogy: Imagine having two joystick directions; span is everywhere you can move your character by pushing those sticks in different amounts.
🔧 Technical explanation: For vectors v,w, span{v,w}={av+bw:a,b∈R}; for example, with v=(10) and w=(01), a=3,b=−2 gives 3v+(−2)w=(3−2) inside the span.
💡 Why it matters: Span tells you the total “territory” your selected vectors can cover. If the span is small (like a line), your vectors can’t reach every point you might care about.
📝 Concrete example: In 2D, two non-collinearvectors span the plane. In 3D, two non-parallel vectorsspan a plane, and three well-chosen vectors span all of 3D; for instance, 2100+3010+(−1)001=23−1 shows a point in 3D spanned by the standard basis.
Shapes of spans:
In 2D, if two vectors are collinear, the span is a line: span{(11),(22)} is just the line y=x; taking a=3,b=−1 gives (11).
In 2D, if two vectors are not collinear, the span is the whole plane: e.g., span{(10),(01)} gives all (xy) via xi^+yj^; for x=2,y=−1 you get (2−1).
In 3D, two non-parallel vectorsspan a plane through the origin: a100+b010=ab0; with a=1,b=2 this is 120.
In 3D, three independent vectors span the full space: x100+y010+z001=xyz; with x=2,y=3,z=4 gives 234.
How to test if a vector is in a span (computational recipe):
Step 1: Write av+bw=u; for example, a(3−2)+b(12)=(40).
Step 2: Match coordinates to get equations: 3a+1b=4 and −2a+2b=0; solving yields a=1,b=1.
Step 3: If a solution exists, u lies in the span; we confirm 1⋅(3−2)+1⋅(12)=(40).
Linear independence
🎯 One-line definition: Vectors are linearly independent if no vector in the set can be formed by combining the others.
🏠 Everyday analogy: Think of team members with different skills; if one person’s skills can be fully replaced by a mix of others, that person is redundant (dependent). If not, the team is independent.
🔧 Technical explanation: For vectors v1,…,vk, independence means the only way a1v1+⋯+akvk=0 happens is when all a1=⋯=ak=0; for example, with v1=(10), v2=(01), the zero vector(00) is achieved only at a1=a2=0.
💡 Why it matters: Independence prevents redundancy and guarantees unique coordinates in a basis. Without independence, your span has more vectors than necessary and coordinate descriptions are not unique.
📝 Concrete example: In 2D, any three vectors are dependent. For instance, (22)=72(3−2)+712(12) shows one vector is a combination of the other two.
Checking independence in practice:
Solve av+bw=0; e.g., for v=(10), w=(01), the equations force a=0,b=0.
In 3D, check if one vector equals a combination of the others: is u=pv+qw? For instance, with u=110, v=100, w=010, we have u=1⋅v+1⋅w, so the set is dependent.
If any nonzero solution exists for coefficients, the set is dependent.
Basis
🎯 One-line definition: A basis is a set of independent vectors that spans the whole space.
🏠 Everyday analogy: It’s like a minimal toolbox where every tool is needed and, together, they can build anything in that workshop.
🔧 Technical explanation: In 2D, any two independent vectors form a basis; in 3D, any three independent vectors form a basis. For example, in 2D, {(3−2),(12)} is a basis because any (xy) can be written as a(3−2)+b(12) for some a,b; try (40) to find a=1,b=1.
💡 Why it matters: A basis gives unique coordinates and the simplest description of every vector in the space. It’s the foundation for understanding matrices and transformations.
📝 Concrete example: The standard basis in 3D, {100,010,001}, can reach any point xyz using x,y,z as coefficients; for example, 2,3,−1 give 23−1.
Dimension
🎯 One-line definition: Dimension is the number of vectors in a smallest basis for a space.
🏠 Everyday analogy: It’s the count of independent directions you need to move freely in that world—like how a character in a 2D game moves left-right and up-down (2 directions), while in 3D they add forward-back (3 directions).
🔧 Technical explanation: 2D spaces have dimension 2 because two independent vectors are necessary and sufficient; 3D spaces have dimension 3 for the same reason. More than the dimension’s worth of vectors must be dependent; for instance, any three vectors in 2D are dependent.
💡 Why it matters: Dimension summarizes complexity and limits. It tells you how many basisvectors you need and when redundancy must happen.
📝 Concrete example: In 2D, (10),(01) form a basis, but adding (22) makes the set dependent because (22)=2⋅(11) where (11)=(10)+(01) gives a concrete dependency: (22)=2⋅(10)+2⋅(01).
Practical procedures derived from the lesson
A) Expressing a vector using a given basis in 2D:
Suppose your basis is b1=(3−2) and b2=(12), and you want coordinates(a,b) for u=(40). Set up ab1+bb2=u, i.e., a(3−2)+b(12)=(40). This gives 3a+b=4 and −2a+2b=0, which solve to a=1,b=1. Thus u=1⋅b1+1⋅b2.
B) Determining if two 2D vectors form a basis:
Check if they are multiples. If v=kw for some number k, they are collinear, not a basis; e.g., (22)=2⋅(11), so these two fail to be a basis. If not multiples, they form a basis for 2D.
C) Understanding planes from spans in 3D:
Two non-parallel vectorsv,w in 3D define a plane {av+bw:a,b∈R}; for example, with v=100 and w=010, you get all points ab0 such as 2−10 when a=2,b=−1.
D) Recognizing dependence in 3D triples:
If u=pv+qw, then {u,v,w} is dependent. For instance, 110=1⋅100+1⋅010 shows dependence.
E) Why bases matter for coordinates:
With a basis{b1,b2} in 2D, every vectoru has a unique pair (a,b) such that u=ab1+bb2; for a check, the earlier example gave u=(40) as 1b1+1b2. If two different pairs worked, their difference would create a nonzero combination equaling zero, contradicting independence.
Tips and warnings
Don’t assume two vectors span the plane just because they look different—verify they are not multiples. For instance, (22) and (11) are different but collinear; any combination like 3(11)−1(22)=(11) stays on the same line.
In 3D, two vectors can’t span the whole space; you need three independent ones, such as 100, 010, 001, which can form 23−1 from 2,3,−1 as coefficients.
Remember that spans through linear combinations always pass through the origin because vectors start at the origin; e.g., a(10)+b(01)=(ab) includes the origin when a=b=0.
If you add more vectors than the dimension, dependence must appear, meaning some vector can be written from the others. In 2D, any triple like (3−2), (12), (22) is necessarily dependent; we computed (22)=72(3−2)+712(12) as a concrete instance.
Putting it all together
Start with basic arrows (vectors) you trust, like the standard basis. Use linear combinations to reach new points, e.g., 2i^+3j^=(23). Look at the entire set of places you can reach; that’s the span. Check whether your set is redundant (dependent) or minimal and complete (a basis). Count your minimal basis size to name the dimension: 2 for the plane, 3 for space.
05Conclusion
This lesson tied together three core building blocks of linear algebra: linear combinations, span, and basisvectors. Linear combinations are the rules of the game—scale vectors and add them—to build any new vector you need; for example, 2(10)+3(01)=(23). The span is the entire region you can reach with those combinations, like a line for collinearvectors, a plane for two independent vectors in 3D, or all of space for three independent 3D vectors; e.g., a100+b010+c001=abc shows how the standard basisspans 3D with a concrete case a=2,b=3,c=−1 giving 23−1. Linear independence ensures there is no redundancy—no vector in your set can be created from the others—which guarantees unique coordinates in a basis and keeps your descriptions clean and minimal. A basis is the perfect balance: just enough independent vectors to cover the entire space, no more and no less. The dimension is simply how many vectors you need in such a basis: 2 for the plane and 3 for space.
To practice, try writing random vectors as combinations of (3,−2) and (1,2), solve small systems like 3a+b=4, −2a+2b=0 to find coefficients, and verify independence by checking whether only the zero coefficients make the zero vector. Build geometric intuition by sketching spans: draw two 2D vectors that are not multiples and shade the whole plane you can reach, then try two that are multiples and watch the span collapse to a line. In 3D, pick two non-parallel vectors and picture the plane they form through the origin, then add a third independent vector to see how the span expands to all of space. These exercises make the algebra and the geometry reinforce each other.
The next logical step after mastering these ideas is to study matrices as machines that move basisvectors and, therefore, move every vector by linearity. Understanding how matrices act on a basis will clarify matrix multiplication, transformations like rotations and scalings, and the meaning of determinants and rank later on. Keep the core message in mind: linear combinations are the language, span is the world they describe, bases are the minimal dictionaries, and dimension is the word count of that dictionary. With these tools, you are ready to describe and navigate any linear space you encounter.
✓Dimension limits independence: in an $n$-dimensional space, any set with more than $n$ vectors is dependent. In 2D, any 3 vectors are dependent; for example, solve to write one as a combination of the other two. In 3D, any 4 are dependent. Use this as a quick check before doing algebra.
✓When picking a basis, verify both spanning and independence. Not being multiples is necessary in 2D; for example, $\begin{pmatrix}3\\-2\end{pmatrix}$ and $\begin{pmatrix}1\\2\end{pmatrix}$ are fine and can reach $\begin{pmatrix}4\\0\end{pmatrix}$ via $a=b=1$. Avoid sets where one can be written from the others.
✓Coordinates in a basis are unique—use this to check your work. If you find two different coefficient pairs for the same vector, you’ve made an error or your set is dependent. For example, $\begin{pmatrix}4\\0\end{pmatrix}$ in the basis $\{\begin{pmatrix}3\\-2\end{pmatrix},\begin{pmatrix}1\\2\end{pmatrix}\}$ must be $1,1$ only. This keeps solutions consistent.
✓Geometric sketches build strong intuition. Draw two 2D vectors that are not multiples, then shade everything you can reach; you’ll see the whole plane. Try two that are multiples and notice it collapses to a line, like multiples of $\begin{pmatrix}1\\1\end{pmatrix}$. Visuals make the algebra feel natural.
✓Always include the origin when thinking about spans. Since linear combinations allow zero coefficients, the origin is always reachable. For example, $0\cdot\begin{pmatrix}1\\0\end{pmatrix} + 0\cdot\begin{pmatrix}0\\1\end{pmatrix} = \begin{pmatrix}0\\0\end{pmatrix}$. This distinguishes spans and subspaces from shifted lines or planes.
✓Use small numeric examples to test ideas quickly. Take simple vectors like $\begin{pmatrix}1\\0\end{pmatrix}$ and $\begin{pmatrix}0\\1\end{pmatrix}$ to check definitions. Compute a few combinations, like $2\hat{i}+(-1)\hat{j}=\begin{pmatrix}2\\-1\end{pmatrix}$. This practice cements understanding.
✓Be careful with look-alike vectors: different does not mean independent. Vectors can look different yet be multiples, like $\begin{pmatrix}2\\2\end{pmatrix}$ and $\begin{pmatrix}1\\1\end{pmatrix}$. Test independence by solving or checking for scalar multiples. This prevents overestimating your span.
✓Tie algebra to geometry: equations describe movements in space. Solving $a\mathbf{v}+b\mathbf{w}=\mathbf{u}$ finds how far to go along each direction arrow to land at $\mathbf{u}$. For instance, for $\begin{pmatrix}4\\0\end{pmatrix}$ with $\begin{pmatrix}3\\-2\end{pmatrix}$ and $\begin{pmatrix}1\\2\end{pmatrix}$, $a=b=1$ maps directly to the path. This dual view is powerful.
✓Expect planes from two 3D directions, and full space from three. Use the standard 3D basis to test examples, like forming $\begin{pmatrix}2\\3\\-1\end{pmatrix}$ with $2,3,-1$. If a third vector lies in the plane of the first two (is a combination), the span stays a plane. That’s a quick diagnostic of dependence.
✓Keep the core trio straight: linear combinations (how you build), span (what you can reach), basis (minimal independent builders). Each plays a unique role but fits together tightly. Understanding them prepares you for matrices and linear transformations. It’s the foundation for all of linear algebra.
Standard basis (2D)
These are the simplest 2D vectors that point along the axes: $\hat{i}=\begin{pmatrix}1\\0\end{pmatrix}$ and $\hat{j}=\begin{pmatrix}0\\1\end{pmatrix}$. Any 2D vector can be written as $x\hat{i}+y\hat{j}$. They make coordinates easy to read. They are independent and span the entire plane.
Standard basis (3D)
These are the three axis-aligned unit vectors in 3D: $\begin{pmatrix}1\\0\\0\end{pmatrix}$, $\begin{pmatrix}0\\1\\0\end{pmatrix}$, and $\begin{pmatrix}0\\0\\1\end{pmatrix}$. Any 3D vector $\begin{pmatrix}x\\y\\z\end{pmatrix}$ can be written using them. They are independent and span 3D space. They give the usual x, y, z coordinates.
Collinear
Vectors are collinear if they lie on the same line through the origin. That means one is a scalar multiple of the other. Collinear vectors cannot span a plane in 2D, only a line. They point in the same or exact opposite directions.
Plane through the origin
In 3D, a plane through the origin is a flat, 2D sheet that includes the origin. It can be described as the span of two independent vectors. Any point on it can be written as $a\mathbf{v}+b\mathbf{w}$. If either vector is changed to be a multiple of the other, it collapses to a line.
