Canonical: https://knowlpedia.com/convex-analysis/linearly-independent-and-linearly-dependent-sets/

# Linear independence and dependence

A set is linearly independent if only the trivial finite linear combination equals zero

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

The set \(M\) is **linearly independent** if for every finite list of distinct vectors \(x_1,\dots,x_m\in M\) and every choice of scalars \(\alpha_1,\dots,\alpha_m\in K\),
\[
\sum_{i=1}^m \alpha_i x_i = 0 \quad\Longrightarrow\quad \alpha_1=\cdots=\alpha_m=0.
\]

If \(M\) is not linearly independent, it is **linearly dependent**, i.e., there exists a finite subset and scalars, not all zero, giving a zero [linear combination](https://knowlpedia.com/convex-analysis/linear-combination/).

## Relation to bases

Linear independence is one of the two defining properties of a [basis](https://knowlpedia.com/convex-analysis/basis-hamel-basis-and-dimension/).

## Examples

- The standard unit vectors \(e_1,\dots,e_n\) in \(\mathbb{R}^n\) are linearly independent.
- The set \(\{(1,0),(2,0)\}\subset\mathbb{R}^2\) is linearly dependent since \(2(1,0)-(2,0)=0\).
- Any set containing the zero vector is linearly dependent: if \(0\in M\), then \(1\cdot 0=0\) is a nontrivial dependence.
- The empty set is linearly independent (there is no finite nonempty subset to witness dependence).
