Canonical: https://knowlpedia.com/convex-analysis/affine-hull-affine-combination/

# Affine Hull and Affine Combination

The smallest affine set containing Ω, and linear combinations with coefficients summing to 1.

Let \(X\) be a [vector space](https://knowlpedia.com/linear-algebra/vector-space/) and let \(\Omega\subset X\).

The **affine hull** of \(\Omega\) is the intersection of all [affine sets](https://knowlpedia.com/convex-analysis/affine-set/) containing \(\Omega\):
\[
\operatorname{aff}(\Omega):=\bigcap\{C\subset X\mid C\text{ is affine and }\Omega\subset C\}.
\]

A vector \(x\in X\) is an **affine combination** of \(\omega_1,\dots,\omega_m\in X\) if
\[
x=\sum_{i=1}^m \lambda_i\omega_i
\quad\text{with}\quad
\sum_{i=1}^m \lambda_i=1.
\]

## Examples

- If \(m=2\) and \(\lambda_1=\lambda\), \(\lambda_2=1-\lambda\), then \(x=\lambda\omega_1+(1-\lambda)\omega_2\) parameterizes the line through \(\omega_1,\omega_2\) as \(\lambda\) varies over \(\mathbb{R}\).
- In \(\mathbb{R}^n\), \(\operatorname{aff}(\Omega)\) is the smallest affine subspace containing \(\Omega\).

## Remarks

Affine combinations are the natural building blocks of affine sets, just as [convex combinations](https://knowlpedia.com/convex-analysis/convex-combination/) are for convex sets.
