0
$\begingroup$

I am currently learning about vector spaces, spaning and dimensions for the first time and am reading old posts on here. I came across this one: If $\dim(V) = n$, is every spaning set $\{v_1,v_2,\ldots,v_n\}$ a basis for $V$?

In it, it is mentioned that "If $\dim V = n$, then any set of $n$ linearly independent vectors will span $V$". I understand that any spanning set of $n$ vectors must consist of linearly independent vectors due to $\dim V = n$ but I don't quite get the other direction.

Could someone help me understand this better?

$\endgroup$
3
  • $\begingroup$ How many subspaces of maximal (finite) dimension can exist? $\endgroup$ Commented Jan 30, 2021 at 20:55
  • $\begingroup$ Can you show (i) every basis of a finite-dimensional vector space has the same size and (ii) $n+1$ vectors in an $n$-dimensional space are linearly dependent? $\endgroup$
    – KCd
    Commented Jan 30, 2021 at 20:58
  • $\begingroup$ Note that a spanning set can be redundant (have more vectors than necessary). A basis is a minimal spanning set. A linearly independent set can be insufficient (cover a subspace rather than the whole space). A basis is also a maximal linearly independent set. We can rightly call this set with the dual properties of being "minimal spanning" and "maximal linearly independent" a basis, because we can prove that (in a vector space) the two properties are equivalent. You perhaps need to review what you already know and test these two aspects. $\endgroup$ Commented Jan 30, 2021 at 21:03

4 Answers 4

1
$\begingroup$

Every linear independent family can be extended to a basis. By definition of $\dim$, this basis still has $n$ members, i.e., we did not really extend our family and already had a basis.

$\endgroup$
1
$\begingroup$

The dimension is defined as the cardinality of a maximal linearly independent set, or the minimal generating set. These two are shown to be the same in any textbook. We say basis to refer to these sets; being both maximal linearly independent and minimal generating sets.

So a set of $\dim V$ linearly independent vectors is a basis basically by definition

$\endgroup$
1
  • 2
    $\begingroup$ The last assertion is true only in finite dimension. $\endgroup$
    – Bernard
    Commented Jan 30, 2021 at 23:03
1
$\begingroup$

Let $\{v_1,\ldots,v_r\}\subset V$ be a set of $r$ linearly independent vectors, and let $\{w_1,\ldots,w_n\}$ be a basis for $V$. Then there exist $\alpha_{ij}$ with $i\in\{1,\ldots,r\}$ and $j\in\{1,\ldots,n\}$ such that $v_i=\sum_{j=1}^n \alpha_{ij}w_j$. Consider the system of equations: $$\sum_{i=1}^r \alpha_{ij} x_i=0, \ j\in\{1,\ldots,n\}$$ Let $(x_1,\ldots,x_n)$ be a solution. Then $$\sum_{i=1}^r x_i v_i=\sum_{i=1}^r\sum_{j=1}^n\alpha_{ij}x_i w_j=\sum_{j=1}^n\left(\sum_{i=1}^r\alpha_{ij}x_i\right)w_j=\sum_{j=1}^n 0\cdot w_j=0$$ $\{v_1,\ldots,v_r\}$ is linearly independent, so it must be $(x_1,\ldots,x_n)=0$. Hence, the previous system of equations either has unique solution or hasn't solution at all. Anyways, this means that it must have at least as many equations as variables, i.e. $r\leq n$.

Then, any linearly independent set must have, at most, $n$ vectors. Let now $\{v_1,\ldots,v_n\}$ be a linearly independent set and let's prove it must span $V$. Assume it doesn't, and take $v\in V\setminus\operatorname{Span}(v_1,\ldots,v_n)$. Then $\{v,v_1,\ldots,v_n\}$ is a linearly independent set with $n+1$ elements; a contradiction. Hence $\{v_1,\ldots,v_n\}$ spans $V$ so it's a basis.

$\endgroup$
2
  • $\begingroup$ i don't get the last part "so it must be $(x_1,...,x_n) = 0$". why does each $x_i$ have to equal $0$? I know that $\{v_1,...,v_r\}$ are linearly independent, I just can't seem to connect the two statements together. $\endgroup$ Commented Oct 6, 2021 at 21:15
  • $\begingroup$ is each $x_i$ a coefficient? or a vector? $\endgroup$ Commented Oct 7, 2021 at 15:25
0
$\begingroup$

Assume that V is a vector space and that you have n linearly independent vectors from it. The dimension of the sub space spanned by the n vectors is n. However since dim V =n, this must be V itself.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .