Change of Basis
In the standard basis , the vector means "".
But if you change to another basis (say and ), the same vector has different coordinates — expressed as a combination of and .
The same point:
in the standard basis: v = (3, 2)
in the new basis: v = 2.5·b₁ + 0.5·b₂ (since 2.5(1,1) + 0.5(1,-1) = (3, 2))
Coordinates are not an intrinsic property of a vector, but a statement “relative to some basis”.
From coordinates in the new basis to coordinates in the standard basis
Let the new basis be (written in coordinates relative to the standard basis). If the coordinates of a vector in the new basis are , then its coordinates in the standard basis are:
The columns of the change-of-basis matrix are the new basis vectors (in coordinates relative to the standard basis).
From coordinates in the standard basis to coordinates in the new basis
Conversely, use :
Example: B = [[1, 1], [1, -1]] (b₁ = (1,1), b₂ = (1,-1))
B⁻¹ = (1/-2)[[-1, -1], [-1, 1]] = [[0.5, 0.5], [0.5, -0.5]]
Check: B⁻¹·(3, 2) = (0.5×3+0.5×2, 0.5×3−0.5×2) = (2.5, 0.5) ✓
(consistent with 2.5b₁ + 0.5b₂ above)
The problem
The same linear transformation is represented by the matrix in the standard basis and by the matrix in the new basis . What is the relation between and ?
The derivation
- First convert the vector from the new-basis coordinates back to the standard basis:
- Apply the transformation:
- Convert back to the new-basis coordinates:
and are said to be similar.
Geometric Intuition
“Representing the same transformation in a different basis” = “go back to the standard view → transform → switch back to the new view”. It is like describing the same thing in a different language — the transformation itself has not changed, only the coordinate system has.
| Scenario | Significance |
|---|---|
| Choosing a “good” basis | under an eigenbasis some transformations become the simplest possible diagonal matrix |
| Diagonalisation | if you find a set of eigenvectors to serve as a basis, the matrix becomes diagonal (the next section) |
| Signal processing | the Fourier transform = changing to the frequency basis |
| Data science | PCA (principal component analysis) = changing to the directions of greatest variance (the eigenvectors) |
The benefit of diagonalisation: computation with a diagonal matrix is extremely simple (element by element), and it also makes taking powers easy:
| Concept | Explanation |
|---|---|
| Coordinates | a statement relative to some basis, not an intrinsic property |
| The change-of-basis matrix | its columns are the new basis vectors in standard-basis coordinates |
| Converting back to the standard basis | |
| Converting to the new basis | |
| Similar matrices | (the same transformation, different coordinate systems) |
| The purpose | to choose a “good basis” that makes the matrix simpler (e.g. diagonalisation) |
Symbols Used in This Article
| Symbol | Type | Pronunciation/Explanation | Meaning in This Article |
|---|---|---|---|
| Mathematical symbol | calligraphic B | The chosen basis (e.g. the new basis ) | |
| Mathematical symbol | b one, b two | The new basis vectors, written in standard-basis coordinates | |
| Mathematical symbol | v in basis B | The coordinates of the vector in the basis , such as | |
| Mathematical symbol | matrix B | The change-of-basis matrix, whose columns are the new basis vectors | |
| Mathematical symbol | B inverse | The inverse of the change-of-basis matrix, converting standard-basis coordinates back to the new basis | |
| Mathematical symbol | A prime equals B inverse A B | The matrix of the same linear transformation in the new basis; and are similar | |
| Greek letter | Lambda | The eigenvalues appearing in the diagonal standard form after diagonalisation |
