Linear span explained

In mathematics, the linear span (also called the linear hull^[1] or just span) of a set of vectors (from a vector space), denoted,^[2] is defined as the set of all linear combinations of the vectors in .^[3] For example, two linearly independent vectors span a plane.The linear span can be characterized either as the intersection of all linear subspaces that contain, or as the smallest subspace containing . The linear span of a set of vectors is therefore a vector space itself. Spans can be generalized to matroids and modules.

To express that a vector space is a linear span of a subset, one commonly uses the following phrases—either: spans, is a spanning set of, is spanned/generated by, or is a generator or generator set of .

Definition

Given a vector space over a field, the span of a set of vectors (not necessarily finite) is defined to be the intersection of all subspaces of that contain . is referred to as the subspace spanned by, or by the vectors in . Conversely, is called a spanning set of, and we say that spans .

Alternatively, the span of may be defined as the set of all finite linear combinations of elements (vectors) of, which follows from the above definition.^{[4] [5] [6] [7]}

$$\backslash operatorname(S)\; =\; \backslash left\; \backslash .$$

In the case of infinite, infinite linear combinations (i.e. where a combination may involve an infinite sum, assuming that such sums are defined somehow as in, say, a Banach space) are excluded by the definition; a generalization that allows these is not equivalent.

Examples

The real vector space

R ^{3} has as a spanning set. This particular spanning set is also a basis. If (−1, 0, 0) were replaced by (1, 0, 0), it would also form the canonical basis of R ^{3} .

Another spanning set for the same space is given by, but this set is not a basis, because it is linearly dependent.

The set is not a spanning set of

R ^{3} , since its span is the space of all vectors in R ^{3} whose last component is zero. That space is also spanned by the set, as (1, 1, 0) is a linear combination of (1, 0, 0) and (0, 1, 0). Thus, the spanned space is not R ^{3.} It can be identified with R ^{2} by removing the third components equal to zero.

The empty set is a spanning set of, since the empty set is a subset of all possible vector spaces in

R ^{3} , and is the intersection of all of these vector spaces.

The set of monomials, where is a non-negative integer, spans the space of polynomials.

Theorems

Equivalence of definitions

The set of all linear combinations of a subset of, a vector space over, is the smallest linear subspace of containing .

Proof. We first prove that is a subspace of . Since is a subset of, we only need to prove the existence of a zero vector in, that is closed under addition, and that is closed under scalar multiplication. Letting

S = \{ v _{1,} v _{2,} \ldots, v _{n} \} , it is trivial that the zero vector of exists in, since 0 = 0 v _{1} + 0 v _{2} + … + 0 v _{n} . Adding together two linear combinations of also produces a linear combination of : ( λ _{1} v _{1} + … + λ _{n} v _{n)} + (\mu _{1} v _{1} + … + \mu _{n} v _{n)} = ( λ _{1} + \mu _{1)} v _{1} + … + ( λ _{n} + \mu _{n)} v _{n} , where all λ _{i,} \mu _{i} \in K , and multiplying a linear combination of by a scalar c \in K will produce another linear combination of : c( λ _{1} v _{1} + … + λ _{n} v _{n)} = c λ _{1} v _{1} + … + c λ _{n} v _{n} . Thus is a subspace of .

Suppose that is a linear subspace of containing . It follows that

S \subseteq \operatorname{span} S , since every is a linear combination of (trivially). Since is closed under addition and scalar multiplication, then every linear combination λ _{1} v _{1} + … + λ _{n} v _{n} must be contained in . Thus, is contained in every subspace of containing, and the intersection of all such subspaces, or the smallest such subspace, is equal to the set of all linear combinations of .

Size of spanning set is at least size of linearly independent set

Every spanning set of a vector space must contain at least as many elements as any linearly independent set of vectors from .

Proof. Let

S = \{ v _{1,} \ldots, v _{m} \} be a spanning set and W = \{ w _{1,} \ldots, w _{n} \} be a linearly independent set of vectors from . We want to show that m \geq n .

Since spans, then

S \cup \{ w _{1} \} must also span, and w _{1} must be a linear combination of . Thus S \cup \{ w _{1} \} is linearly dependent, and we can remove one vector from that is a linear combination of the other elements. This vector cannot be any of the, since is linearly independent. The resulting set is \{ w _{1,} v _{1,} \ldots, v _{i-1} , v _{i+1} , \ldots, v _{m} \} , which is a spanning set of . We repeat this step times, where the resulting set after the th step is the union of \{ w _{1,} \ldots, w _{p} \} and vectors of .

It is ensured until the th step that there will always be some to remove out of for every adjoint of, and thus there are at least as many 's as there are 's—i.e.

m \geq n . To verify this, we assume by way of contradiction that m < n . Then, at the th step, we have the set \{ w _{1,} \ldots, w _{m} \} and we can adjoin another vector w _{m+1} . But, since \{ w _{1,} \ldots, w _{m} \} is a spanning set of, w _{m+1} is a linear combination of \{ w _{1,} \ldots, w _{m} \} . This is a contradiction, since is linearly independent.

Spanning set can be reduced to a basis

Let be a finite-dimensional vector space. Any set of vectors that spans can be reduced to a basis for, by discarding vectors if necessary (i.e. if there are linearly dependent vectors in the set). If the axiom of choice holds, this is true without the assumption that has finite dimension. This also indicates that a basis is a minimal spanning set when is finite-dimensional.

Generalizations

Generalizing the definition of the span of points in space, a subset of the ground set of a matroid is called a spanning set if the rank of equals the rank of the entire ground set

The vector space definition can also be generalized to modules.^{[8] [9]} Given an -module and a collection of elements, ..., of, the submodule of spanned by, ..., is the sum of cyclic modules$$Ra\_1\; +\; \backslash cdots\; +\; Ra\_n\; =\; \backslash left\backslash$$consisting of all R-linear combinations of the elements . As with the case of vector spaces, the submodule of A spanned by any subset of A is the intersection of all submodules containing that subset.

Closed linear span (functional analysis)

In functional analysis, a closed linear span of a set of vectors is the minimal closed set which contains the linear span of that set.

Suppose that is a normed vector space and let be any non-empty subset of . The closed linear span of, denoted by

\overline{\operatorname{Sp}}(E) or \overline{\operatorname{Span}}(E) , is the intersection of all the closed linear subspaces of which contain .

One mathematical formulation of this is

\overline{\operatorname{Sp}}(E) = \{u\in X | \forall\varepsilon > 0 \exists x\in\operatorname{Sp}(E) : \|x - u\|<\varepsilon\}.

The closed linear span of the set of functions xⁿ on the interval [0, 1], where n is a non-negative integer, depends on the norm used. If the L² norm is used, then the closed linear span is the Hilbert space of square-integrable functions on the interval. But if the maximum norm is used, the closed linear span will be the space of continuous functions on the interval. In either case, the closed linear span contains functions that are not polynomials, and so are not in the linear span itself. However, the cardinality of the set of functions in the closed linear span is the cardinality of the continuum, which is the same cardinality as for the set of polynomials.

Notes

The linear span of a set is dense in the closed linear span. Moreover, as stated in the lemma below, the closed linear span is indeed the closure of the linear span.

Closed linear spans are important when dealing with closed linear subspaces (which are themselves highly important, see Riesz's lemma).

A useful lemma

Let be a normed space and let be any non-empty subset of . Then

(So the usual way to find the closed linear span is to find the linear span first, and then the closure of that linear span.)

See also

• Affine hull
• Conical combination
• Convex hull

Citations

1. . Linear Hull.
2. pp. 29-30, §§ 2.5, 2.8
3. p. 29, § 2.7
4. p. 100, ch. 2, Definition 2.13
5. pp. 29-30, §§ 2.5, 2.8
6. pp. 41-42
7. Vector Space Span.
8. p. 96, ch. 4
9. p. 193, ch. 6

Sources

Textbooks

• Book: Axler, Sheldon Jay. Linear Algebra Done Right. Springer. 2015. 978-3-319-11079-0. 3rd. Sheldon Axler.
• Book: Hefferon, Jim. Linear Algebra. Orthogonal Publishing. 2020. 978-1-944325-11-4. 4th. Jim Hefferon.
• Book: Mac Lane. Saunders. Algebra. Birkhoff. Garrett. AMS Chelsea Publishing. 1999. 978-0821816462. 3rd. Saunders Mac Lane. Garrett Birkhoff. 1988.
• Book: Oxley, James G. . James Oxley

. James Oxley . 2nd . 9780199202508 . Oxford University Press . Oxford Graduate Texts in Mathematics . Matroid Theory . 3 . 2011.

• Book: Roman, Steven. Advanced Linear Algebra. Springer. 2005. 0-387-24766-1. 2nd. Steven Roman.
• Book: Rynne. Brian P.. Linear Functional Analysis. Youngson. Martin A.. Springer. 2008. 978-1848000049.
• Lay, David C. (2021) Linear Algebra and Its Applications (6th Edition). Pearson.

Web

• Web site: Lankham. Isaiah. Nachtergaele. Bruno. Bruno Nachtergaele. Schilling. Anne. Anne Schilling. 13 February 2010. Linear Algebra - As an Introduction to Abstract Mathematics. 27 September 2011. University of California, Davis.
• Web site: Weisstein. Eric Wolfgang. Eric W. Weisstein. Vector Space Span. 16 Feb 2021. MathWorld. CITEREFMathWorld2021.
• Web site: 5 April 2020. Linear hull. 16 Feb 2021. Encyclopedia of Mathematics. CITEREFEncyclopedia_of_Mathematics2020.

External links

• Linear Combinations and Span: Understanding linear combinations and spans of vectors, khanacademy.org.
• Web site: Linear combinations, span, and basis vectors. Essence of Linear Algebra. Sanderson. Grant. 3Blue1Brown. August 6, 2016. YouTube. https://ghostarchive.org/varchive/youtube/20211211/k7RM-ot2NWY. 2021-12-11. live.