Solved Exercises on Coordinates with Respect to a Basis

Master coordinate systems, basis representations, and change of basis through these 5 detailed exercises.

Coordinates with Respect to a Basis - Comprehensive Summary

Definitions

Coordinates with Respect to a Basis: If B = {v₁, v₂, ..., vₙ} is a basis for vector space V, then for any vector x ∈ V, there exist unique scalars c₁, c₂, ..., cₙ such that x = c₁v₁ + c₂v₂ + ... + cₙvₙ. The coordinates of x with respect to B are [x]_B = [c₁, c₂, ..., cₙ]ᵀ.

Standard Basis: The natural basis for Rⁿ where eᵢ has 1 in the i-th position and 0 elsewhere.

Change of Basis: Converting coordinates from one basis to another using transformation matrices.

Core Rules

• Coordinates are unique for each basis

• The coordinate vector represents the linear combination weights

• Change of basis matrices are invertible

• Dimension of coordinate vector equals dimension of space

Methods

1. Set up the equation: x = c₁v₁ + c₂v₂ + ... + cₙvₙ

2. Convert to matrix equation: [v₁ v₂ ... vₙ][c₁, c₂, ..., cₙ]ᵀ = x

3. Solve for the coordinate vector [c₁, c₂, ..., cₙ]ᵀ

4. Use the inverse matrix for change of basis

Key Tips

• Always verify that the given vectors form a basis

• Use row reduction for solving systems efficiently

• Remember that coordinates depend on the order of basis vectors

• Check your work by reconstructing the original vector

Solution: Exercises 1 to 3
1 Basic Coordinate Calculation
Exercise 1
Let \(B = \left\{ \begin{pmatrix} 1 \\ 0 \end{pmatrix}, \begin{pmatrix} 1 \\ 1 \end{pmatrix} \right\}\) be a basis for \(\mathbb{R}^2\). Find the coordinates of \(\vec{x} = \begin{pmatrix} 3 \\ 2 \end{pmatrix}\) with respect to B.
Definition:

Coordinates with Respect to a Basis: If B = {v₁, v₂, ..., vₙ} is a basis for vector space V, then for any vector x ∈ V, the coordinates of x with respect to B are the unique scalars [c₁, c₂, ..., cₙ]ᵀ such that x = c₁v₁ + c₂v₂ + ... + cₙvₙ.

Coordinate Calculation Method:
  1. Set up the equation: \(\vec{x} = c_1\vec{v_1} + c_2\vec{v_2} + \ldots + c_n\vec{v_n}\)
  2. Form the matrix equation: \([v_1 \, v_2 \, \ldots \, v_n]\begin{pmatrix} c_1 \\ c_2 \\ \vdots \\ c_n \end{pmatrix} = \vec{x}\)
  3. Solve for the coordinate vector \([c_1, c_2, \ldots, c_n]^T\)
  4. Express the result as \([\vec{x}]_B\)
Basis Matrix
\(P = \begin{pmatrix} 1 & 1 \\ 0 & 1 \end{pmatrix}\)
System
\(P\begin{pmatrix} c_1 \\ c_2 \end{pmatrix} = \begin{pmatrix} 3 \\ 2 \end{pmatrix}\)
Solution
\([\vec{x}]_B = \begin{pmatrix} 1 \\ 2 \end{pmatrix}\)
Step 1: Verify that B is a basis

Check that the vectors are linearly independent:

\(a\begin{pmatrix} 1 \\ 0 \end{pmatrix} + b\begin{pmatrix} 1 \\ 1 \end{pmatrix} = \begin{pmatrix} 0 \\ 0 \end{pmatrix}\)

This gives us: \(\begin{pmatrix} a + b \\ b \end{pmatrix} = \begin{pmatrix} 0 \\ 0 \end{pmatrix}\)

So: \(b = 0\) and \(a + b = 0\), which means \(a = 0\).

Since the only solution is \(a = b = 0\), the vectors are linearly independent. ✓

Step 2: Set up the equation

We want to find scalars \(c_1\) and \(c_2\) such that:

\(\vec{x} = c_1\begin{pmatrix} 1 \\ 0 \end{pmatrix} + c_2\begin{pmatrix} 1 \\ 1 \end{pmatrix}\)

\(\begin{pmatrix} 3 \\ 2 \end{pmatrix} = \begin{pmatrix} c_1 + c_2 \\ c_2 \end{pmatrix}\)

Step 3: Solve the system of equations

From the vector equation, we get the system:

\(\begin{cases} c_1 + c_2 = 3 \quad \text{(equation 1)} \\ c_2 = 2 \quad \text{(equation 2)} \end{cases}\)

From equation 2: \(c_2 = 2\)

Substituting into equation 1: \(c_1 + 2 = 3\), so \(c_1 = 1\)

Step 4: Verify the solution

Check: \(1\begin{pmatrix} 1 \\ 0 \end{pmatrix} + 2\begin{pmatrix} 1 \\ 1 \end{pmatrix} = \begin{pmatrix} 1 \\ 0 \end{pmatrix} + \begin{pmatrix} 2 \\ 2 \end{pmatrix} = \begin{pmatrix} 3 \\ 2 \end{pmatrix}\). ✓

Step 5: Write the coordinate vector

The coordinates of \(\vec{x}\) with respect to basis B are:

\([\vec{x}]_B = \begin{pmatrix} 1 \\ 2 \end{pmatrix}\)

\([\vec{x}]_B = \begin{pmatrix} 1 \\ 2 \end{pmatrix}\)
Final answer:

The coordinates of \(\vec{x}\) with respect to basis B are \(\begin{pmatrix} 1 \\ 2 \end{pmatrix}\).

Applied rules:

• Linear independence verification: Essential before coordinate calculation

• System solving: Convert vector equation to scalar system

• Verification: Always check the solution by reconstruction

2 Change of Basis
Exercise 2
Let \(B = \left\{ \begin{pmatrix} 1 \\ 0 \end{pmatrix}, \begin{pmatrix} 0 \\ 1 \end{pmatrix} \right\}\) and \(C = \left\{ \begin{pmatrix} 1 \\ 1 \end{pmatrix}, \begin{pmatrix} 1 \\ -1 \end{pmatrix} \right\}\) be two bases for \(\mathbb{R}^2\). Find the change-of-basis matrix from B to C.
Definition:

Change-of-Basis Matrix: The matrix P_C←B transforms coordinates from basis B to basis C. If [v]_B are the coordinates of vector v in basis B, then [v]_C = P_C←B [v]_B.

Basis Matrices
\(P_B = \begin{pmatrix} 1 & 0 \\ 0 & 1 \end{pmatrix}\), \(P_C = \begin{pmatrix} 1 & 1 \\ 1 & -1 \end{pmatrix}\)
Inverse
\(P_C^{-1} = \begin{pmatrix} 1/2 & 1/2 \\ 1/2 & -1/2 \end{pmatrix}\)
Change Matrix
\(P_{C \leftarrow B} = \begin{pmatrix} 1/2 & 1/2 \\ 1/2 & -1/2 \end{pmatrix}\)
Step 1: Form the basis matrices

\(P_B = \begin{pmatrix} 1 & 0 \\ 0 & 1 \end{pmatrix}\) (standard basis matrix)

\(P_C = \begin{pmatrix} 1 & 1 \\ 1 & -1 \end{pmatrix}\) (matrix with C basis vectors as columns)

Step 2: Find the inverse of P_C

For \(P_C = \begin{pmatrix} 1 & 1 \\ 1 & -1 \end{pmatrix}\), we have:

\(\det(P_C) = (1)(-1) - (1)(1) = -2\)

\(P_C^{-1} = \frac{1}{-2}\begin{pmatrix} -1 & -1 \\ -1 & 1 \end{pmatrix} = \begin{pmatrix} 1/2 & 1/2 \\ 1/2 & -1/2 \end{pmatrix}\)

Step 3: Calculate the change-of-basis matrix

The change-of-basis matrix from B to C is:

\(P_{C \leftarrow B} = P_C^{-1}P_B = P_C^{-1} \cdot I = P_C^{-1}\)

Therefore: \(P_{C \leftarrow B} = \begin{pmatrix} 1/2 & 1/2 \\ 1/2 & -1/2 \end{pmatrix}\)

Step 4: Verification

Let's verify with \(\vec{x} = \begin{pmatrix} 1 \\ 0 \end{pmatrix}\) (which has coordinates \(\begin{pmatrix} 1 \\ 0 \end{pmatrix}\) in B).

In basis C: \(\begin{pmatrix} 1/2 & 1/2 \\ 1/2 & -1/2 \end{pmatrix}\begin{pmatrix} 1 \\ 0 \end{pmatrix} = \begin{pmatrix} 1/2 \\ 1/2 \end{pmatrix}\)

Check: \(\frac{1}{2}\begin{pmatrix} 1 \\ 1 \end{pmatrix} + \frac{1}{2}\begin{pmatrix} 1 \\ -1 \end{pmatrix} = \begin{pmatrix} 1 \\ 0 \end{pmatrix}\). ✓

Step 5: General change-of-basis formula

For any vector \(\vec{v}\): \([\vec{v}]_C = P_{C \leftarrow B}[\vec{v}]_B\)

\(P_{C \leftarrow B} = \begin{pmatrix} 1/2 & 1/2 \\ 1/2 & -1/2 \end{pmatrix}\)
Final answer:

The change-of-basis matrix from B to C is \(\begin{pmatrix} 1/2 & 1/2 \\ 1/2 & -1/2 \end{pmatrix}\).

Applied rules:

• Change matrix formula: \(P_{C \leftarrow B} = P_C^{-1}P_B\)

• Inverse calculation: Use determinant and adjugate method

• Verification: Test with known vectors to confirm correctness

3 3D Coordinates
Exercise 3
Let \(B = \left\{ \begin{pmatrix} 1 \\ 0 \\ 0 \end{pmatrix}, \begin{pmatrix} 1 \\ 1 \\ 0 \end{pmatrix}, \begin{pmatrix} 1 \\ 1 \\ 1 \end{pmatrix} \right\}\) be a basis for \(\mathbb{R}^3\). Find the coordinates of \(\vec{x} = \begin{pmatrix} 2 \\ 3 \\ 4 \end{pmatrix}\) with respect to B.
Definition:

3D Coordinates: For a basis B = {v₁, v₂, v₃} in \(\mathbb{R}^3\), the coordinates of vector x with respect to B are the unique scalars [c₁, c₂, c₃]ᵀ such that x = c₁v₁ + c₂v₂ + c₃v₃.

System
\(c_1 + c_2 + c_3 = 2\), \(c_2 + c_3 = 3\), \(c_3 = 4\)
Solution
\(c_1 = 1\), \(c_2 = -1\), \(c_3 = 4\)
Coordinate Vector
\([\vec{x}]_B = \begin{pmatrix} 1 \\ -1 \\ 4 \end{pmatrix}\)
Step 1: Verify that B is a basis

We need to check that the three vectors are linearly independent:

\(a\begin{pmatrix} 1 \\ 0 \\ 0 \end{pmatrix} + b\begin{pmatrix} 1 \\ 1 \\ 0 \end{pmatrix} + c\begin{pmatrix} 1 \\ 1 \\ 1 \end{pmatrix} = \begin{pmatrix} 0 \\ 0 \\ 0 \end{pmatrix}\)

This gives us: \(\begin{pmatrix} a + b + c \\ b + c \\ c \end{pmatrix} = \begin{pmatrix} 0 \\ 0 \\ 0 \end{pmatrix}\)

From the last equation: \(c = 0\)

From the middle equation: \(b + c = 0 \Rightarrow b = 0\)

From the first equation: \(a + b + c = 0 \Rightarrow a = 0\)

Since the only solution is \(a = b = c = 0\), the vectors are linearly independent. ✓

Step 2: Set up the equation

We want to find scalars \(c_1\), \(c_2\), and \(c_3\) such that:

\(\vec{x} = c_1\begin{pmatrix} 1 \\ 0 \\ 0 \end{pmatrix} + c_2\begin{pmatrix} 1 \\ 1 \\ 0 \end{pmatrix} + c_3\begin{pmatrix} 1 \\ 1 \\ 1 \end{pmatrix}\)

\(\begin{pmatrix} 2 \\ 3 \\ 4 \end{pmatrix} = \begin{pmatrix} c_1 + c_2 + c_3 \\ c_2 + c_3 \\ c_3 \end{pmatrix}\)

Step 3: Solve the system of equations

From the vector equation, we get the system:

\(\begin{cases} c_1 + c_2 + c_3 = 2 \quad \text{(equation 1)} \\ c_2 + c_3 = 3 \quad \text{(equation 2)} \\ c_3 = 4 \quad \text{(equation 3)} \end{cases}\)

From equation 3: \(c_3 = 4\)

Substituting into equation 2: \(c_2 + 4 = 3\), so \(c_2 = -1\)

Substituting into equation 1: \(c_1 + (-1) + 4 = 2\), so \(c_1 = -1\)

Step 4: Verify the solution

Check: \((-1)\begin{pmatrix} 1 \\ 0 \\ 0 \end{pmatrix} + (-1)\begin{pmatrix} 1 \\ 1 \\ 0 \end{pmatrix} + 4\begin{pmatrix} 1 \\ 1 \\ 1 \end{pmatrix}\)

\(= \begin{pmatrix} -1 \\ 0 \\ 0 \end{pmatrix} + \begin{pmatrix} -1 \\ -1 \\ 0 \end{pmatrix} + \begin{pmatrix} 4 \\ 4 \\ 4 \end{pmatrix} = \begin{pmatrix} 2 \\ 3 \\ 4 \end{pmatrix}\). ✓

Step 5: Write the coordinate vector

The coordinates of \(\vec{x}\) with respect to basis B are:

\([\vec{x}]_B = \begin{pmatrix} -1 \\ -1 \\ 4 \end{pmatrix}\)

\([\vec{x}]_B = \begin{pmatrix} -1 \\ -1 \\ 4 \end{pmatrix}\)
Final answer:

The coordinates of \(\vec{x}\) with respect to basis B are \(\begin{pmatrix} -1 \\ -1 \\ 4 \end{pmatrix}\).

Applied rules:

• Back-substitution: Solve 3D systems by back-substitution when possible

• Linear independence: Always verify basis conditions

• Verification: Reconstruct the original vector to confirm solution

Coordinates with Respect to a Basis Properties
\([\vec{x}]_B = P_B^{-1}\vec{x}\)
Coordinate Formula
Change of Basis
\([\vec{x}]_C = P_{C \leftarrow B}[\vec{x}]_B\)
Transform coordinates between bases
Uniqueness
If B is basis, then [x]_B is unique
Coordinates are uniquely determined
Linearity
\([c_1\vec{x_1} + c_2\vec{x_2}]_B = c_1[\vec{x_1}]_B + c_2[\vec{x_2}]_B\)
Coordinate mapping is linear
Coordinate Mapping: The mapping x ↦ [x]_B is a linear transformation from V to R^n
Dimension Preservation: The coordinate vector has the same dimension as the basis size
Isomorphism: Vector spaces of the same dimension are isomorphic via coordinate mapping
Solution: Exercises 4 to 5
4 Polynomial Space Coordinates
Exercise 4
Let \(B = \{1, x, x^2\}\) be the standard basis for the space of polynomials of degree ≤ 2. Find the coordinates of \(p(x) = 3x^2 - 2x + 5\) with respect to B.
Definition:

Polynomial Coordinates: For a polynomial space with basis B = {b₁(x), b₂(x), ..., bₙ(x)}, the coordinates of polynomial p(x) with respect to B are the unique scalars [c₁, c₂, ..., cₙ]ᵀ such that p(x) = c₁b₁(x) + c₂b₂(x) + ... + cₙbₙ(x).

Standard Basis
\(B = \{1, x, x^2\}\)
Polynomial
\(p(x) = 5 \cdot 1 + (-2) \cdot x + 3 \cdot x^2\)
Coordinate Vector
\([p(x)]_B = \begin{pmatrix} 5 \\ -2 \\ 3 \end{pmatrix}\)
Step 1: Identify the basis

The standard basis for polynomials of degree ≤ 2 is \(B = \{1, x, x^2\}\).

Step 2: Express the polynomial in terms of the basis

We have: \(p(x) = 3x^2 - 2x + 5\)

Reordering by degree: \(p(x) = 5 \cdot 1 + (-2) \cdot x + 3 \cdot x^2\)

Step 3: Identify the coefficients

The coefficients of the basis elements are:

Coefficient of 1: 5

Coefficient of x: -2

Coefficient of x²: 3

Step 4: Form the coordinate vector

The coordinate vector with respect to B is formed by arranging the coefficients in order:

\([p(x)]_B = \begin{pmatrix} 5 \\ -2 \\ 3 \end{pmatrix}\)

Step 5: Verification

Check: \(5 \cdot 1 + (-2) \cdot x + 3 \cdot x^2 = 5 - 2x + 3x^2 = p(x)\). ✓

Step 6: Alternative basis example

If we had a different basis, say \(C = \{1, 1+x, 1+x+x^2\}\), we would need to solve a system to find the coordinates.

\([p(x)]_B = \begin{pmatrix} 5 \\ -2 \\ 3 \end{pmatrix}\)
Final answer:

The coordinates of \(p(x) = 3x^2 - 2x + 5\) with respect to the standard basis B are \(\begin{pmatrix} 5 \\ -2 \\ 3 \end{pmatrix}\).

Applied rules:

• Standard basis: Coefficients directly become coordinates

• Polynomial representation: Write polynomial as linear combination of basis elements

• Order matters: Coordinate vector order matches basis element order

5 Inverse Change of Basis
Exercise 5
Given the change-of-basis matrix \(P_{C \leftarrow B} = \begin{pmatrix} 1 & 2 \\ 3 & 4 \end{pmatrix}\) from basis B to basis C, find the change-of-basis matrix from C to B.
Definition:

Inverse Change of Basis: If \(P_{C \leftarrow B}\) transforms coordinates from basis B to basis C, then \(P_{B \leftarrow C} = (P_{C \leftarrow B})^{-1}\) transforms coordinates from basis C to basis B.

Given Matrix
\(P_{C \leftarrow B} = \begin{pmatrix} 1 & 2 \\ 3 & 4 \end{pmatrix}\)
Determinant
\(\det = 1 \cdot 4 - 2 \cdot 3 = -2\)
Inverse Matrix
\(P_{B \leftarrow C} = \begin{pmatrix} -2 & 1 \\ 3/2 & -1/2 \end{pmatrix}\)
Step 1: Understand the relationship

If \([\vec{v}]_C = P_{C \leftarrow B}[\vec{v}]_B\), then \([\vec{v}]_B = (P_{C \leftarrow B})^{-1}[\vec{v}]_C\).

Therefore: \(P_{B \leftarrow C} = (P_{C \leftarrow B})^{-1}\).

Step 2: Calculate the determinant

For \(P_{C \leftarrow B} = \begin{pmatrix} 1 & 2 \\ 3 & 4 \end{pmatrix}\):

\(\det(P_{C \leftarrow B}) = (1)(4) - (2)(3) = 4 - 6 = -2\)

Step 3: Find the adjugate matrix

For a 2×2 matrix \(\begin{pmatrix} a & b \\ c & d \end{pmatrix}\), the adjugate is \(\begin{pmatrix} d & -b \\ -c & a \end{pmatrix}\).

Adjugate of \(P_{C \leftarrow B}\) is \(\begin{pmatrix} 4 & -2 \\ -3 & 1 \end{pmatrix}\).

Step 4: Calculate the inverse

\(P_{B \leftarrow C} = (P_{C \leftarrow B})^{-1} = \frac{1}{\det(P_{C \leftarrow B})} \cdot \text{adjugate}(P_{C \leftarrow B})\)

\(= \frac{1}{-2}\begin{pmatrix} 4 & -2 \\ -3 & 1 \end{pmatrix} = \begin{pmatrix} -2 & 1 \\ 3/2 & -1/2 \end{pmatrix}\)

Step 5: Verify the result

Check that \(P_{C \leftarrow B} \cdot P_{B \leftarrow C} = I\):

\(\begin{pmatrix} 1 & 2 \\ 3 & 4 \end{pmatrix}\begin{pmatrix} -2 & 1 \\ 3/2 & -1/2 \end{pmatrix}\)

\(= \begin{pmatrix} 1(-2) + 2(3/2) & 1(1) + 2(-1/2) \\ 3(-2) + 4(3/2) & 3(1) + 4(-1/2) \end{pmatrix}\)

\(= \begin{pmatrix} -2 + 3 & 1 - 1 \\ -6 + 6 & 3 - 2 \end{pmatrix} = \begin{pmatrix} 1 & 0 \\ 0 & 1 \end{pmatrix} = I\). ✓

Step 6: Interpretation

If \([\vec{v}]_C = \begin{pmatrix} a \\ b \end{pmatrix}\) are the coordinates of vector \(\vec{v}\) in basis C, then the coordinates in basis B are:

\([\vec{v}]_B = \begin{pmatrix} -2 & 1 \\ 3/2 & -1/2 \end{pmatrix}\begin{pmatrix} a \\ b \end{pmatrix} = \begin{pmatrix} -2a + b \\ (3/2)a - (1/2)b \end{pmatrix}\)

\(P_{B \leftarrow C} = \begin{pmatrix} -2 & 1 \\ 3/2 & -1/2 \end{pmatrix}\)
Final answer:

The change-of-basis matrix from C to B is \(\begin{pmatrix} -2 & 1 \\ 3/2 & -1/2 \end{pmatrix}\).

Applied rules:

• Inverse relationship: \(P_{B \leftarrow C} = (P_{C \leftarrow B})^{-1}\)

• 2×2 inverse: Use determinant and adjugate formula

• Verification: Check that product equals identity matrix

Coordinates with Respect to a Basis: Advanced Concepts
\([\vec{x}]_B = \begin{pmatrix} c_1 \\ c_2 \\ \vdots \\ c_n \end{pmatrix} \text{ where } \vec{x} = c_1\vec{v_1} + c_2\vec{v_2} + \ldots + c_n\vec{v_n}\)
Coordinate Definition
Key definitions:

Coordinate Vector: The coordinate vector of x with respect to basis B represents x in the coordinate system defined by B.

Change of Basis: The process of converting coordinates from one basis to another using transformation matrices.

Transition Matrix: The matrix that maps coordinates in one basis to coordinates in another basis.

Complete methodology:
  1. For coordinate calculation: Solve the linear system formed by expressing x as a linear combination of basis vectors
  2. For change of basis: Use the formula \(P_{C \leftarrow B} = P_C^{-1}P_B\)
  3. For verification: Reconstruct the original vector using the coordinates and basis vectors
  4. For inverse transformations: Use the inverse of the change-of-basis matrix
Tip 1: Always verify that your basis vectors are linearly independent before calculating coordinates.
Tip 2: The order of basis vectors matters in the coordinate vector.
Tip 3: Use row reduction for solving large systems of equations efficiently.
Tip 4: Check your work by reconstructing the original vector from coordinates.
Common errors: Miscounting basis vectors, forgetting to verify linear independence, confusing coordinate vector with original vector, incorrect order in change-of-basis matrices.
Applications: Computer graphics (coordinate systems), physics (reference frames), engineering (coordinate transformations), machine learning (feature transformations).
Properties to know by heart:

• Coordinates are unique for each basis

• The coordinate mapping is linear: \([c_1\vec{u} + c_2\vec{v}]_B = c_1[\vec{u}]_B + c_2[\vec{v}]_B\)

• Change-of-basis matrices are always invertible

• \(P_{B \leftarrow C} = (P_{C \leftarrow B})^{-1}\)

• \([\vec{x}]_B = P_B^{-1}\vec{x}\) when B is a basis matrix

Coordinates Visualization: Basis Change Effect
Exercise 6: Coordinate Systems
Visualize how the same vector \(\vec{x} = \begin{pmatrix} 3 \\ 1 \end{pmatrix}\) has different coordinates in the standard basis B and an alternative basis C. Show how the coordinate system changes with the basis.

Geometric Interpretation: The same vector appears at different coordinates depending on the basis used for measurement. The vector itself remains fixed in space, but its representation changes.

  • Standard basis B: \(\left\{\begin{pmatrix} 1 \\ 0 \end{pmatrix}, \begin{pmatrix} 0 \\ 1 \end{pmatrix}\right\}\)
  • Alternative basis C: \(\left\{\begin{pmatrix} 1 \\ 1 \end{pmatrix}, \begin{pmatrix} 1 \\ -1 \end{pmatrix}\right\}\)
  • Same vector, different coordinate representations

Questions & Answers

Question: I'm confused about why we need different coordinate systems. Why can't we just use the standard basis all the time?

Answer: Great question! While the standard basis is convenient, different bases offer several advantages:

Computational simplicity:

  • Some problems become simpler in specific bases (e.g., diagonalization)
  • Certain transformations have simpler representations
  • Numerical stability can improve with appropriate bases

Physical relevance:

  • In physics, natural coordinate systems align with problem symmetries
  • In computer graphics, object-local coordinates are more intuitive
  • In data science, principal component basis reveals important patterns

Think of it like measuring temperature - Celsius and Fahrenheit measure the same thing but in different scales. Similarly, different bases measure the same vector in different "scales" that may be more appropriate for specific purposes.

Question: How do I know if a set of vectors forms a valid basis?

Answer: A set of vectors forms a basis if and only if it satisfies two conditions:

1. Linear Independence:

  • The only solution to \(c_1\vec{v_1} + c_2\vec{v_2} + \ldots + c_n\vec{v_n} = \vec{0}\) is \(c_1 = c_2 = \ldots = c_n = 0\)
  • Check by forming a matrix with the vectors as columns and computing its rank
  • Alternatively, check if the determinant is non-zero (for square matrices)

2. Spanning:

  • Every vector in the space can be expressed as a linear combination of the basis vectors
  • For \(\mathbb{R}^n\), if you have n linearly independent vectors, they automatically span the space
  • Check by verifying that the matrix has full rank

For an n-dimensional space, any set of n linearly independent vectors automatically forms a basis.

Question: What's the relationship between the change-of-basis matrix and the coordinate transformation?

Answer: The relationship is fundamental:

Change-of-basis matrix: If \(P_{C \leftarrow B}\) is the change-of-basis matrix from basis B to basis C, then:

\([\vec{v}]_C = P_{C \leftarrow B}[\vec{v}]_B\)

How it works:

  • \([\vec{v}]_B\) is the coordinate vector of \(\vec{v}\) in basis B
  • \([\vec{v}]_C\) is the coordinate vector of \(\vec{v}\) in basis C
  • \(P_{C \leftarrow B}\) transforms coordinates from B-system to C-system

The change-of-basis matrix is computed as \(P_{C \leftarrow B} = P_C^{-1}P_B\), where \(P_B\) and \(P_C\) are the matrices whose columns are the basis vectors of B and C respectively.

This means: "To find coordinates in C-basis, multiply B-coordinates by the change-of-basis matrix."

Question: How are coordinates with respect to a basis used in physics?

Answer: Coordinate systems with respect to different bases are essential in physics:

Reference frames:

  • Changing from Earth frame to rotating frame (Coriolis forces)
  • Special relativity: Lorentz transformations between reference frames
  • Quantum mechanics: Changing between different observable bases

Coordinate systems:

  • Cartesian to polar/cylindrical/spherical coordinates
  • Body-fixed vs. space-fixed coordinate systems
  • Principal axes of rotation for rigid bodies

Quantum mechanics:

  • Wave functions in position vs. momentum basis
  • Spin states in different measurement directions
  • Change of basis corresponds to measurement in different observables

The mathematical framework of coordinate transformations provides the foundation for understanding how physical laws appear in different reference frames.

Question: How do coordinate systems relate to machine learning and data science?

Answer: Coordinate systems and basis changes are fundamental in machine learning:

Dimensionality reduction:

  • PCA changes to principal component basis (eigenvectors)
  • Data is represented in new coordinate system with maximum variance
  • Reduces noise and captures essential features

Feature engineering:

  • Linear combinations of features create new coordinate systems
  • Fourier transforms change to frequency domain basis
  • Wavelet transforms use wavelet basis

Neural networks:

  • Each layer applies a coordinate transformation
  • Weights define the transformation matrix
  • Activation functions introduce non-linearity

Understanding coordinate systems helps in feature selection, model interpretability, and algorithm design.