Linear Dependence

Student project for MA265



Contents

Definition




The vectors v1,v2,...,vk in a vector space V are said to be linearly dependent if there exist constans a1,a2,...,ak not all zero, such that

                                                                           $ \sum_{j=1}^k a_1v_1+a_2v_2+...+a_kv_k=0 $                       (1)                                 

Otherwise, v1,v2,...,vk are called linearly independent. That is v1,v2,...,vk are linearly independent if, whenever a1v1+ a2v2+...+akvk=0


                                                                       a1= a2=...=ak=0



If S = {v1,v2,...,vk} ,then we also say that the set S is linearly dependent or linearly independent if the vectors have the corresponding property.


         It should be emphasized that for any vectors v1,v2,...,vk, Equation (1) always holds if we choose all the scalars a1,a2,...,ak, equal to zero. The important point in this definition is whether it is possible to satisfy (1) with at least one of the scalars different from zero.


Remark: Definition is started for a finite set of vectors, but it also applies to an infinite set S of a vector space, using corresponding notation for infinite sums.


        To determine whether a set of vectors is linearly independent or linearly dependent, we use Equation (1). Regardless of the form of the vectors, Equation (1) yields a homogeneous linear system of equations. it is always consistent, since a1=a2=...=ak=0 is a solution. However, the main idea from the Definitionis whether there is a nontrivial solution.





Example 1



Determine whether the vectors

$ v1=\left[\begin{array}{cccc}3\\2\\1\end{array}\right] $

$ v2=\left[\begin{array}{cccc}1\\2\\0\end{array}\right] $

$ v3=\left[\begin{array}{cccc}-1\\-2\\-1\end{array}\right] $

are linearly independent.


Solution

Forming Equation (1)

$ a1\left[\begin{array}{cccc}3\\2\\1\end{array}\right] $$ +a2\left[\begin{array}{cccc}1\\2\\0\end{array}\right] $$ +a3\left[\begin{array}{cccc}-1\\2\\-1\end{array}\right] $$ =\left[\begin{array}{cccc}0\\0\\0\end{array}\right] $

we obtain the homogeneous system

3a1 + a2a3 = 0

2a1 + 2a2 + 2a3 = 0

a1               − a3 = 0


The corresponding augmented matrix is

$ \left[\begin{array}{cccc}3&1&-1&|0\\2&2&2&|0\\1&0&-1&|0\end{array}\right] $


Whose reduced row echelon form is

$ \left[\begin{array}{cccc}1&0&-1&|0\\0&1&2&|0\\0&0&0&|0\end{array}\right] $


This there is a nontrivial solution

$ \left[\begin{array}{cccc}k\\-2k\\k\end{array}\right] $       k is not equal to 0

so the vectors are linearly dependent.






Example 2



Are the vectors

$ v1=\left[\begin{array}{cccc}1&0&1&2\end{array}\right] $

$ v2=\left[\begin{array}{cccc}0&1&1&2\end{array}\right] $

$ v3=\left[\begin{array}{cccc}1&1&1&3\end{array}\right] $     in R4

linearly dependent or linearly independent?


Solution

We form Equation (1),

a1v1+a2v2+a3v3=0

and solve for a1,a2,a3.

The resulting homogeneous system is

a1 + a3 = 0

a2 + a3 = 0

a1 + a2 + a3 = 0

2a1 + 2a2 + 3a3 = 0


The corresponding augmented matrix is

$ \left[\begin{array}{cccc}1&0&1&|0\\0&1&1&|0\\1&1&1&|0\\2&2&3&|0\end{array}\right] $


and its reduced row echelon form is

$ \left[\begin{array}{cccc}1&0&0&|0\\0&1&0&|0\\0&0&1&|0\\0&0&0&|0\end{array}\right] $


Thus the only solution is the trivial solution

a1 = a2 = a3 = 0

so the vectors are linearly independent.



Theorem

Let {S = v1,v2,...,vn}

be a set of n vectors in Rn. Let A be the matrix whose columns (rows) are the elements of S. Then S is linearly independent if and only if det(A) is not equal to 0.



Example 3


Is     $ S=\left[\begin{array}{cccc}1&2&3\end{array}\right] $ $ ,\left[\begin{array}{cccc}0&1&2\end{array}\right] $ $ ,\left[\begin{array}{cccc}3&0&-1\end{array}\right] $

a linearly independent set of vectors in R3?


Solution

We form the matrix A whose rows are the vectors in S:


$ A=\left[\begin{array}{cccc}1&2&3\\0&1&2\\3&0&-1\end{array}\right] $


Since det(A) = 2, we conclude that S is linearly independent.



Theorem

Let S1 and S2 be finite susbsets of a vector space and let S1 be a subset of S2. Then the following statements are true:

(a) If S1 is linearly dependent, so is S2.

(b) If S2 is linearly independent, so is S1.



Theorem

The nonzero vectors v1,v2,...,vn in a vector space V are linearly dependent if and only if one of the vectors vj $ (2 \le j) $ is a linear combination of the preceding vectors v1,v2,...,vj − 1.


Example 4


Let V=R3 and also 

$ v1=\left[\begin{array}{cccc}1&2&-1\end{array}\right] $

$ v2=\left[\begin{array}{cccc}1&-2&1\end{array}\right] $

$ v3=\left[\begin{array}{cccc}-3&2&-1\end{array}\right] $

$ v4=\left[\begin{array}{cccc}2&0&0\end{array}\right] $


We find that

                                            v1 + v2 + 0v3v4 = 0,

so v!,v2,v3, and v

are linearly dependent. We then have

                                             v4 = v1 + v2 + 0v3


Remarks

1. We observe that the previous Theorem does not say that every vector v is a linear combination of the preceding vectors. Thus, in the previous example, we also have v1 + 2v2 + v3 + 0v4 = 0.

We cannot solve, in this equation, for v4 as a linear combination of

v1,v2, and v3, since its coefficient is zero.

2. We can also prove that if S = {v1,v2,...,vk} is a set of vectors in a vector space V, then S is linearly dependent if and only if one of the vectors in S is a linear combination of all the other vectors in S. For instance in the previous example,

                                                   v1 = − v2 − 0v3 + v4;      $ v_2= \frac{-1}{2} v1 - \frac{1}{2} v_3 -0v_4 $

3. Observe that if v1,v2,...,vk

are linearly independent vectors in a vector space, then they must be distinct and nonzero.




Useful link to calculate linear dependence/independence

http://www.math.odu.edu/~bogacki/cgi-bin/lat.cgi?c=li

http://www.math.odu.edu/~bogacki/cgi-bin/lat.cgi


Alumni Liaison

Ph.D. on Applied Mathematics in Aug 2007. Involved on applications of image super-resolution to electron microscopy

Francisco Blanco-Silva