Basis, dimension, and vector spaces
A basis is a minimal set of building blocks for a space. The number of vectors in any basis, the dimension, measures how big the space is.
Vector spaces
A vector space is any set of objects that can be added and scaled while obeying the usual rules: closure, commutativity, associativity, a zero element, additive inverses, and distributive laws. Beyond R^n, examples include polynomials of degree at most n, m x n matrices, continuous functions, and solutions of linear differential equations. Linear algebra's results apply to all of them.
Subspaces
A subspace is a subset of a vector space that is itself a vector space: it contains the zero vector and is closed under addition and scalar multiplication. In R^3, the subspaces are the origin alone, lines through the origin, planes through the origin, and R^3 itself. A plane not through the origin is not a subspace, since it lacks the zero vector.
Basis
A basis of a vector space is a set of vectors that is linearly independent and spans the space. Every vector can then be written in exactly one way as a combination of basis vectors. The standard basis of R^n is e_1, ..., e_n; the polynomials of degree at most 2 have the basis 1, x, x^2. A space has many different bases.
Dimension
All bases of a given space contain the same number of vectors, called its dimension. R^n has dimension n; polynomials of degree at most 2 form a space of dimension 3; 2 x 2 matrices form a space of dimension 4. The zero space has dimension 0, and spaces such as all continuous functions are infinite-dimensional.
Coordinates
Given a basis \(B = \{\mathbf{b}_1, \dots, \mathbf{b}_n\}\), the coordinates of \(\mathbf{x}\) are the unique weights with \(\mathbf{x} = c_1\mathbf{b}_1 + \dots + c_n\mathbf{b}_n\), written \([\mathbf{x}]_B\). Coordinates turn any n-dimensional space into a copy of \(\mathbb{R}^n\): the polynomial \(2 - x + 5x^2\) has coordinates (2, -1, 5) in the basis 1, x, x^2. Changing basis changes coordinates, not the vector itself.
Useful theorems
In an n-dimensional space, any n independent vectors automatically form a basis, as do any n vectors that span; any independent set can be extended to a basis; any spanning set contains a basis; and any set with more than n vectors is dependent. These results save work: to show n vectors form a basis of R^n, checking independence alone is enough.
Why dimension matters
Dimension counts degrees of freedom: the number of independent directions or parameters. A plane through the origin in R^3 has dimension 2 because points on it need two coordinates. In data science, dimensionality reduction seeks lower-dimensional subspaces that capture most of the variation in high-dimensional data.
Worked example
Is \(\{(1, 1, 0), (0, 1, 1), (1, 0, 1)\}\) a basis of \(\mathbb{R}^3\)? Row reducing the matrix with these columns gives pivots in all three columns, since its determinant is \(1(1 - 0) - 0 + 1(1 - 0) = 2 \ne 0\). Three independent vectors in a 3-dimensional space form a basis. The coordinates of (2, 3, 1) solve \(c_1 + c_3 = 2\), \(c_1 + c_2 = 3\), \(c_2 + c_3 = 1\), giving \(c = (2, 1, 0)\).
Key takeaways
Vector spaces are sets closed under addition and scaling with the usual rules: R^n, polynomials, matrices, and functions. Subspaces contain zero and are closed under both operations; in R^3 they are the origin, lines and planes through the origin, and R^3. A basis is an independent spanning set, giving unique coordinates. Every basis of a space has the same size, its dimension: n for R^n, 3 for quadratics, 4 for 2 x 2 matrices. In dimension n, n independent vectors or n spanning vectors form a basis. Dimension counts degrees of freedom.
Flashcard
Set closed under addition and scaling with the usual rules
Check your understanding
What is the dimension of the space of polynomials of degree at most 3?
