01 Vectors and combos

Definition, basic operations

A sequence of n elements of \(f\), v, is a vector.

Addition, scalar multiplication

Addition of vectors is naturally defined: \((u + v)_i = u_i + v_i\). Scalar multiplication is similarly defined.

Geometric model: coefficient sequence

Can be interpreted as a point in the cartesian space \(F^{n}\); but which point? That depends on what the basis vectors model, the units, the inner product between them.

As a combination of standard basis vectors

Define standard basis vectors: \(e_i\) as a vector with 1 in the ith position, 0 elesewhere.

So, v can be viewed as a combination \(v_i e_i\) of the standard basis vectors \(\set{e_i}\).

Standard basis models what?

The standard basis are usually considered with the standard inner product: \(\dprod{x, y} = \sum_i x_i y_i\). But this need not be the case. Other bases are possible.

Equivalent representations of the same point

So, the vector representing a point in a coordinate space changes, when the basis chosen is different. Change of basis operation is a linear operation, for details see linear algebra ref.

Combinations of vectors

Linear combination

\(\sum a_{i} v_{i} = p\).

Conic/ non-ve combination

Coefficients \(a_{i} \geq 0\).

Affine combination

If coefficients \(\sum a_{i} = 1\), this is an affine combination.

Colleniearity preserved

Make affine combo \(p = ax + (1-a)y\); take vector \(x - p = (1-a)(x-y)\); this has same inner product with the basis vectors as x-y.

Convex combination

Affine combo where \(a_{i} \geq 0\): both affine and conic.