Solved Exercises on Transition Matrices

Master transition matrices: change of basis, coordinate transformations, and properties of transition matrices through these 5 detailed exercises.

Solution: Exercises 1 to 3
1 Basic Transition Matrix
Exercise 1
Find the transition matrix from basis B = {(1, 2), (3, 4)} to basis C = {(1, 0), (0, 1)} (the standard basis). Then use it to convert [v]_B = \(\begin{bmatrix} 2 \\ 1 \end{bmatrix}\) to [v]_C.
Definition:

Transition matrix: P_{C←B} such that [v]_C = P_{C←B}[v]_B

Method:
  1. Form matrices B and C with basis vectors as columns
  2. Calculate P_{C←B} = C⁻¹B
  3. Apply transformation: [v]_C = P_{C←B}[v]_B
Basis matrices
B = \(\begin{bmatrix} 1 & 3 \\ 2 & 4 \end{bmatrix}\), C = I
Transition matrix
P_{C←B} = B
Step 1: Form the basis matrices

B = \(\begin{bmatrix} 1 & 3 \\ 2 & 4 \end{bmatrix}\) (columns are vectors in basis B)

C = \(\begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix}\) = I (standard basis is identity matrix)

Step 2: Calculate the transition matrix

Since C = I, we have C⁻¹ = I

Therefore: P_{C←B} = C⁻¹B = IB = B

So P_{C←B} = \(\begin{bmatrix} 1 & 3 \\ 2 & 4 \end{bmatrix}\)

Step 3: Apply the transformation

[v]_C = P_{C←B}[v]_B

[v]_C = \(\begin{bmatrix} 1 & 3 \\ 2 & 4 \end{bmatrix} \begin{bmatrix} 2 \\ 1 \end{bmatrix}\)

= \(\begin{bmatrix} (1)(2) + (3)(1) \\ (2)(2) + (4)(1) \end{bmatrix} = \begin{bmatrix} 5 \\ 8 \end{bmatrix}\)

Step 4: Verify by converting back to vector form

In standard coordinates: v = 5(1, 0) + 8(0, 1) = (5, 8)

Check using B coordinates: v = 2(1, 2) + 1(3, 4) = (2, 4) + (3, 4) = (5, 8) ✓

Step 5: Generalize the result

For any basis B to standard basis C, P_{C←B} is simply the matrix with B's vectors as columns

This is because the standard basis vectors are the "reference" for coordinate representation

P_{C←B} = \(\begin{bmatrix} 1 & 3 \\ 2 & 4 \end{bmatrix}\), [v]_C = \(\begin{bmatrix} 5 \\ 8 \end{bmatrix}\)
Final answer:

The transition matrix is P_{C←B} = \(\begin{bmatrix} 1 & 3 \\ 2 & 4 \end{bmatrix}\) and [v]_C = \(\begin{bmatrix} 5 \\ 8 \end{bmatrix}\)

Applied rules:

• Transition formula: [v]_C = P_{C←B}[v]_B

• Standard basis: Transition to standard basis uses basis matrix directly

• Verification: Convert back to verify correctness

2 Transition Between Two Non-Standard Bases
Exercise 2
Find the transition matrix from basis B = {(1, 1), (2, 3)} to basis C = {(1, 2), (1, 1)}. Then convert [w]_B = \(\begin{bmatrix} 4 \\ 1 \end{bmatrix}\) to [w]_C.
Definition:

General transition matrix: P_{C←B} = C⁻¹B where B and C are matrices with basis vectors as columns

Basis matrices
B = \(\begin{bmatrix} 1 & 2 \\ 1 & 3 \end{bmatrix}\), C = \(\begin{bmatrix} 1 & 1 \\ 2 & 1 \end{bmatrix}\)
Transition matrix
P_{C←B} = \(\begin{bmatrix} -1 & 1 \\ 3 & -1 \end{bmatrix}\)
Step 1: Form the basis matrices

B = \(\begin{bmatrix} 1 & 2 \\ 1 & 3 \end{bmatrix}\) (columns: (1,1), (2,3))

C = \(\begin{bmatrix} 1 & 1 \\ 2 & 1 \end{bmatrix}\) (columns: (1,2), (1,1))

Step 2: Calculate C⁻¹

For C = \(\begin{bmatrix} 1 & 1 \\ 2 & 1 \end{bmatrix}\): det(C) = (1)(1) - (1)(2) = -1

Using 2×2 inverse formula: C⁻¹ = \(\frac{1}{-1}\begin{bmatrix} 1 & -1 \\ -2 & 1 \end{bmatrix} = \begin{bmatrix} -1 & 1 \\ 2 & -1 \end{bmatrix}\)

Step 3: Calculate P_{C←B} = C⁻¹B

P_{C←B} = \(\begin{bmatrix} -1 & 1 \\ 2 & -1 \end{bmatrix} \begin{bmatrix} 1 & 2 \\ 1 & 3 \end{bmatrix}\)

= \(\begin{bmatrix} (-1)(1) + (1)(1) & (-1)(2) + (1)(3) \\ (2)(1) + (-1)(1) & (2)(2) + (-1)(3) \end{bmatrix}\)

= \(\begin{bmatrix} 0 & 1 \\ 1 & 1 \end{bmatrix}\)

Step 4: Apply the transformation

[w]_C = P_{C←B}[w]_B = \(\begin{bmatrix} 0 & 1 \\ 1 & 1 \end{bmatrix} \begin{bmatrix} 4 \\ 1 \end{bmatrix}\)

= \(\begin{bmatrix} (0)(4) + (1)(1) \\ (1)(4) + (1)(1) \end{bmatrix} = \begin{bmatrix} 1 \\ 5 \end{bmatrix}\)

Step 5: Verify the result

In basis C: w = 1(1,2) + 5(1,1) = (1,2) + (5,5) = (6,7)

In basis B: w = 4(1,1) + 1(2,3) = (4,4) + (2,3) = (6,7) ✓

Both representations yield the same vector in standard coordinates

P_{C←B} = \(\begin{bmatrix} 0 & 1 \\ 1 & 1 \end{bmatrix}\), [w]_C = \(\begin{bmatrix} 1 \\ 5 \end{bmatrix}\)
Final answer:

The transition matrix is P_{C←B} = \(\begin{bmatrix} 0 & 1 \\ 1 & 1 \end{bmatrix}\) and [w]_C = \(\begin{bmatrix} 1 \\ 5 \end{bmatrix}\)

Applied rules:

• General formula: P_{C←B} = C⁻¹B

• Matrix multiplication: Standard algorithm

• Verification: Convert to standard coordinates to check

3 Inverse Transition Matrix
Exercise 3
Using the bases from Exercise 2, find the transition matrix from C to B, i.e., P_{B←C}. Verify that P_{B←C} = (P_{C←B})⁻¹.
Definition:

Inverse property: P_{B←C} = (P_{C←B})⁻¹

From Ex. 2
P_{C←B} = \(\begin{bmatrix} 0 & 1 \\ 1 & 1 \end{bmatrix}\)
Inverse
P_{B←C} = \(\begin{bmatrix} -1 & 1 \\ 1 & 0 \end{bmatrix}\)
Step 1: Find P_{B←C} using the formula

P_{B←C} = B⁻¹C

From Exercise 2: B = \(\begin{bmatrix} 1 & 2 \\ 1 & 3 \end{bmatrix}\), C = \(\begin{bmatrix} 1 & 1 \\ 2 & 1 \end{bmatrix}\)

det(B) = (1)(3) - (2)(1) = 1

B⁻¹ = \(\begin{bmatrix} 3 & -2 \\ -1 & 1 \end{bmatrix}\)

P_{B←C} = B⁻¹C = \(\begin{bmatrix} 3 & -2 \\ -1 & 1 \end{bmatrix} \begin{bmatrix} 1 & 1 \\ 2 & 1 \end{bmatrix}\)

= \(\begin{bmatrix} 3-4 & 3-2 \\ -1+2 & -1+1 \end{bmatrix} = \begin{bmatrix} -1 & 1 \\ 1 & 0 \end{bmatrix}\)

Step 2: Calculate (P_{C←B})⁻¹ directly

From Exercise 2: P_{C←B} = \(\begin{bmatrix} 0 & 1 \\ 1 & 1 \end{bmatrix}\)

det(P_{C←B}) = (0)(1) - (1)(1) = -1

(P_{C←B})⁻¹ = \(\frac{1}{-1}\begin{bmatrix} 1 & -1 \\ -1 & 0 \end{bmatrix} = \begin{bmatrix} -1 & 1 \\ 1 & 0 \end{bmatrix}\)

Step 3: Verify equality

P_{B←C} = \(\begin{bmatrix} -1 & 1 \\ 1 & 0 \end{bmatrix}\) = (P_{C←B})⁻¹ ✓

Step 4: Verify the inverse property

P_{C←B} × P_{B←C} should equal I:

\(\begin{bmatrix} 0 & 1 \\ 1 & 1 \end{bmatrix} \begin{bmatrix} -1 & 1 \\ 1 & 0 \end{bmatrix} = \begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix} = I\) ✓

Step 5: Apply to example from Exercise 2

Given [w]_C = \(\begin{bmatrix} 1 \\ 5 \end{bmatrix}\) (from Exercise 2), let's verify:

[w]_B = P_{B←C}[w]_C = \(\begin{bmatrix} -1 & 1 \\ 1 & 0 \end{bmatrix} \begin{bmatrix} 1 \\ 5 \end{bmatrix} = \begin{bmatrix} -1+5 \\ 1+0 \end{bmatrix} = \begin{bmatrix} 4 \\ 1 \end{bmatrix}\)

This matches [w]_B = \(\begin{bmatrix} 4 \\ 1 \end{bmatrix}\) from Exercise 2 ✓

P_{B←C} = (P_{C←B})⁻¹ = \(\begin{bmatrix} -1 & 1 \\ 1 & 0 \end{bmatrix}\)
Final answer:

The inverse transition matrix is P_{B←C} = \(\begin{bmatrix} -1 & 1 \\ 1 & 0 \end{bmatrix}\)

Applied rules:

• Inverse property: P_{B←C} = (P_{C←B})⁻¹

• Verification: P_{C←B}P_{B←C} = I

• Consistency: Both methods yield the same result

Transition Matrices: Key Concepts
[v]_C = P_{C \leftarrow B}[v]_B
Transition Formula
Formula
P_{C \leftarrow B} = C^{-1}B
Transition Matrix
Property
P_{B \leftarrow C} = (P_{C \leftarrow B})^{-1}
Inverse Property
Property
P_{D \leftarrow A} = P_{D \leftarrow C}P_{C \leftarrow B}P_{B \leftarrow A}
Chain Rule
Invertibility: Transition matrices are always invertible
Identity: P_{B←B} = I for any basis B
Orthogonal basis: For orthonormal bases, P_{C←B} = (P_{B←C})ᵀ
Solution: Exercises 4 to 5
4 Transition and Matrix Powers
Exercise 4
Let A = \(\begin{bmatrix} 4 & -2 \\ 1 & 1 \end{bmatrix}\) and let B = {(1, 1), (2, 1)} be a basis. Find the matrix representation of A with respect to basis B, i.e., [A]_B. Then use it to find A⁴ in the B-coordinate system.
Definition:

Matrix representation: [A]_B = P_{B←E}AP_{E←B} where E is standard basis and P_{E←B} = B

B matrix
B = \(\begin{bmatrix} 1 & 2 \\ 1 & 1 \end{bmatrix}\)
[A]_B
[A]_B = \(\begin{bmatrix} 0 & 2 \\ 3 & 3 \end{bmatrix}\)
Step 1: Form the change of basis matrix

B = \(\begin{bmatrix} 1 & 2 \\ 1 & 1 \end{bmatrix}\) (columns are basis vectors)

B⁻¹ = \(\frac{1}{(1)(1)-(2)(1)}\begin{bmatrix} 1 & -2 \\ -1 & 1 \end{bmatrix} = \frac{1}{-1}\begin{bmatrix} 1 & -2 \\ -1 & 1 \end{bmatrix} = \begin{bmatrix} -1 & 2 \\ 1 & -1 \end{bmatrix}\)

Step 2: Calculate [A]_B = B⁻¹AB

[A]_B = \(\begin{bmatrix} -1 & 2 \\ 1 & -1 \end{bmatrix} \begin{bmatrix} 4 & -2 \\ 1 & 1 \end{bmatrix} \begin{bmatrix} 1 & 2 \\ 1 & 1 \end{bmatrix}\)

First, calculate B⁻¹A:

\(\begin{bmatrix} -1 & 2 \\ 1 & -1 \end{bmatrix} \begin{bmatrix} 4 & -2 \\ 1 & 1 \end{bmatrix} = \begin{bmatrix} (-1)(4)+(2)(1) & (-1)(-2)+(2)(1) \\ (1)(4)+(-1)(1) & (1)(-2)+(-1)(1) \end{bmatrix}\)

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

Step 3: Complete the calculation

[A]_B = \(\begin{bmatrix} -2 & 4 \\ 3 & -3 \end{bmatrix} \begin{bmatrix} 1 & 2 \\ 1 & 1 \end{bmatrix}\)

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

= \(\begin{bmatrix} 2 & 0 \\ 0 & 3 \end{bmatrix}\)

Step 4: Calculate [A⁴]_B

Since [A]_B is diagonal, [A⁴]_B = ([A]_B)⁴

[A⁴]_B = \(\begin{bmatrix} 2 & 0 \\ 0 & 3 \end{bmatrix}^4 = \begin{bmatrix} 2^4 & 0 \\ 0 & 3^4 \end{bmatrix} = \begin{bmatrix} 16 & 0 \\ 0 & 81 \end{bmatrix}\)

Step 5: Verify the diagonalization

The fact that [A]_B is diagonal means the basis B consists of eigenvectors of A

We can verify: A(1,1) = (4,-2) + (1,1) = (2,2) = 2(1,1)

A(2,1) = (8,-4) + (2,1) = (6,3) = 3(2,1)

So (1,1) is eigenvector with eigenvalue 2, and (2,1) is eigenvector with eigenvalue 3

[A]_B = \(\begin{bmatrix} 2 & 0 \\ 0 & 3 \end{bmatrix}\), [A⁴]_B = \(\begin{bmatrix} 16 & 0 \\ 0 & 81 \end{bmatrix}\)
Final answer:

The matrix representation is [A]_B = \(\begin{bmatrix} 2 & 0 \\ 0 & 3 \end{bmatrix}\) and [A⁴]_B = \(\begin{bmatrix} 16 & 0 \\ 0 & 81 \end{bmatrix}\)

Applied rules:

• Matrix representation: [A]_B = B⁻¹AB

• Diagonal matrices: Powers computed element-wise

• Eigenvector basis: Results in diagonal matrix representation

5 Orthogonal Transition Matrix
Exercise 5
Let U = {(1/√2, 1/√2), (-1/√2, 1/√2)} be an orthonormal basis. Find the transition matrix from U to the standard basis E. Show that it's orthogonal and verify P⁻¹ = Pᵀ.
Definition:

Orthogonal matrix: Matrix where PᵀP = PPᵀ = I, so P⁻¹ = Pᵀ

Transition matrix
P_{E←U} = \(\begin{bmatrix} 1/\sqrt{2} & -1/\sqrt{2} \\ 1/\sqrt{2} & 1/\sqrt{2} \end{bmatrix}\)
Verification
PᵀP = I
Step 1: Form the transition matrix

Since we're going from U to E (standard basis), and E is the identity, P_{E←U} is formed by placing the U vectors as columns:

P_{E←U} = \(\begin{bmatrix} 1/\sqrt{2} & -1/\sqrt{2} \\ 1/\sqrt{2} & 1/\sqrt{2} \end{bmatrix}\)

Step 2: Calculate Pᵀ

Pᵀ = \(\begin{bmatrix} 1/\sqrt{2} & 1/\sqrt{2} \\ -1/\sqrt{2} & 1/\sqrt{2} \end{bmatrix}\)

Step 3: Verify PᵀP = I

PᵀP = \(\begin{bmatrix} 1/\sqrt{2} & 1/\sqrt{2} \\ -1/\sqrt{2} & 1/\sqrt{2} \end{bmatrix} \begin{bmatrix} 1/\sqrt{2} & -1/\sqrt{2} \\ 1/\sqrt{2} & 1/\sqrt{2} \end{bmatrix}\)

Row 1, Col 1: \(\frac{1}{\sqrt{2}} \cdot \frac{1}{\sqrt{2}} + \frac{1}{\sqrt{2}} \cdot \frac{1}{\sqrt{2}} = \frac{1}{2} + \frac{1}{2} = 1\)

Row 1, Col 2: \(\frac{1}{\sqrt{2}} \cdot \left(-\frac{1}{\sqrt{2}}\right) + \frac{1}{\sqrt{2}} \cdot \frac{1}{\sqrt{2}} = -\frac{1}{2} + \frac{1}{2} = 0\)

Row 2, Col 1: \(\left(-\frac{1}{\sqrt{2}}\right) \cdot \frac{1}{\sqrt{2}} + \frac{1}{\sqrt{2}} \cdot \frac{1}{\sqrt{2}} = -\frac{1}{2} + \frac{1}{2} = 0\)

Row 2, Col 2: \(\left(-\frac{1}{\sqrt{2}}\right) \cdot \left(-\frac{1}{\sqrt{2}}\right) + \frac{1}{\sqrt{2}} \cdot \frac{1}{\sqrt{2}} = \frac{1}{2} + \frac{1}{2} = 1\)

So PᵀP = \(\begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix} = I\) ✓

Step 4: Verify PPᵀ = I

PPᵀ = \(\begin{bmatrix} 1/\sqrt{2} & -1/\sqrt{2} \\ 1/\sqrt{2} & 1/\sqrt{2} \end{bmatrix} \begin{bmatrix} 1/\sqrt{2} & 1/\sqrt{2} \\ -1/\sqrt{2} & 1/\sqrt{2} \end{bmatrix}\)

Similar calculation shows PPᵀ = I ✓

Step 5: Verify P⁻¹ = Pᵀ

Since PᵀP = I, we have Pᵀ = P⁻¹ by definition of matrix inverse

This confirms that transition matrices between orthonormal bases are orthogonal

P_{E←U} is orthogonal with P⁻¹ = Pᵀ
Final answer:

The transition matrix P_{E←U} is orthogonal, confirming P⁻¹ = Pᵀ

Applied rules:

• Orthogonal matrix: PᵀP = I ⟺ P⁻¹ = Pᵀ

• Orthonormal basis: Transition matrix is orthogonal

• Geometric interpretation: Preserves lengths and angles

Transition Matrices: Comprehensive Guide
[v]_{B'} = P_{B' \leftarrow B}[v]_B
General Transition Formula
Key definitions:

Transition matrix: P_{B'←B} such that [v]_{B'} = P_{B'←B}[v]_B

Coordinate vector: Representation of vector with respect to a basis

Change of basis: Converting coordinate representations between different bases

Complete methodology:
  1. Identify bases: Clearly specify B and B'
  2. Form basis matrices: Create matrices with basis vectors as columns
  3. Calculate transition matrix: P_{B'←B} = (B')⁻¹B
  4. Apply transformation: [v]_{B'} = P_{B'←B}[v]_B
  5. Verify results: Convert back to check consistency
Tip 1: If B' is the standard basis, P_{B'←B} = B (the basis matrix itself).
Tip 2: For orthonormal bases, transition matrices are orthogonal: P⁻¹ = Pᵀ.
Tip 3: P_{B←B'} = (P_{B'←B})⁻¹ - always verify this inverse relationship.
Tip 4: The transition matrix converts coordinates from the right basis to the left basis.
Common errors: Confusing the direction of transition, miscalculating matrix inverses, forgetting to verify results.
Exam preparation: Master 2×2 and 3×3 cases, understand the geometric meaning.
Properties to know by heart:

• Transition property: [v]_{B'} = P_{B'←B}[v]_B

• Inverse property: P_{B←B'} = (P_{B'←B})⁻¹

• Chain rule: P_{C←A} = P_{C←B}P_{B←A}

• Identity: P_{B←B} = I

• Orthogonal: For orthonormal bases, P⁻¹ = Pᵀ

• Composition: Transition matrices compose like functions

Transition Matrix Properties Visualization
Exercise 6: Property Analysis
Visual analysis of transition matrix properties:
• Orthogonality conditions
• Inverse relationships
• Coordinate transformation properties

Analysis: The visualization shows how different transition matrix properties behave.

  • Orthogonal matrices preserve norms
  • Inverse property holds for all valid transitions
  • Chain rule enables complex basis changes

Questions & Answers

Question: Why do we need transition matrices? Can't we just work in the standard basis?

Answer: While the standard basis is convenient, other bases offer significant advantages:

Computational benefits:

  • Diagonalization: Matrices become diagonal in eigenbasis, making powers and exponentials trivial
  • Numerical stability: Some problems are more stable in appropriate coordinate systems
  • Sparsity: Many applications result in sparse matrices in the right basis

Physical/geometric interpretation:

Many problems have natural coordinate systems. For example, in mechanics, principal axes of rotation are more natural than arbitrary coordinates.

Applications:

  • Quantum mechanics: Different bases correspond to different observables
  • Signal processing: Fourier basis for frequency analysis
  • Machine learning: Principal component analysis uses optimal basis

Transition matrices allow us to move between different perspectives of the same mathematical object, enabling us to choose the most convenient representation for each problem.

Question: How do I remember the formula P_{C←B} = C⁻¹B? The subscripts seem confusing.

Answer: Use this memory aid:

Reading direction: P_{Target←Source} converts from Source to Target coordinates

Formula breakdown: P_{C←B} = C⁻¹B

  • C⁻¹: Converts from basis C back to standard coordinates
  • B: Converts from standard coordinates to basis B
  • Together: From B coordinates → standard coordinates → C coordinates

Alternative approach:

Think of it as "going backwards": to go from B to C, we first undo B (with B⁻¹) then apply C (with C⁻¹), but since we want B → standard → C, we use C⁻¹B.

Verification method:

Always check with a simple example: if [v]_B = e₁ (first standard basis vector in B coordinates), then v should be the first vector of B in standard coordinates. Apply P_{C←B} and verify the result makes sense.

Question: What are practical applications of transition matrices?

Answer: Transition matrices have extensive practical applications:

  • Computer graphics: Converting between world, camera, and screen coordinate systems
  • Robotics: Converting between joint space and Cartesian space coordinates
  • Quantum mechanics: Changing between different observable bases
  • Signal processing: Converting between time and frequency domains
  • Data science: Principal component analysis and dimensionality reduction
  • Physics: Converting between different reference frames

Specific examples:

In computer graphics, when rendering a scene, vertices are transformed from model coordinates to world coordinates to camera coordinates to screen coordinates using a chain of transition matrices.

In structural engineering, stress and strain tensors are often converted between different coordinate systems to analyze material behavior along principal directions.

In machine learning, PCA finds a transition matrix to convert high-dimensional data to a lower-dimensional space that preserves the most variance.

The concept is fundamental because it allows us to work in the most convenient coordinate system for each stage of a problem while maintaining mathematical consistency through the transition relationships.