MAT 350: Linear Independence

Dr. Gilbert

August 9, 2025

Warm-Up Problems

Complete the following warm-up problems to re-familiarize yourself with concepts we’ll be leveraging today.

  1. Determine whether the vector \(\vec{b} = \begin{bmatrix} -5\\ -2\\ -1\\ 4\end{bmatrix}\) is in \(\text{span}\left(\left\{\begin{bmatrix} 1\\ 1\\ 0\\ 2\end{bmatrix}, \begin{bmatrix} -3\\ 0\\ 1\\ 1\end{bmatrix}, \begin{bmatrix} 0\\ 1\\ 1\\ 0\end{bmatrix}\right\}\right)\)

  2. Do the columns of the matrix \(\begin{bmatrix} 1 & 8 & -2 & 3\\ 0 & 0 & -1 & 1\\ 0 & 0 & 0 & 4\\ 0 & 0 & 0 & 0\end{bmatrix}\) span \(\mathbb{R}^4\)? Why or why not?

  3. Describe the space spanned by the vectors \(\vec{v_1} = \begin{bmatrix} 1\\ 0\\ -1\end{bmatrix}\), \(\vec{v_2} =\begin{bmatrix} 2\\ 3\\ 0\end{bmatrix}\), and \(\vec{v_3} = \begin{bmatrix} 6\\ 15\\ 4\end{bmatrix}\).

Reminders and Today’s Goal

  • Recently we’ve discussed linear combinations of vectors and spans of collections of vectors.

    • A linear combination of vectors \(\vec{v_1}, \vec{v_2}, \dots, \vec{v_k}\) is any vector that can be written as \(\vec{y} = c_1\vec{v_1} + c_2\vec{v_2} + \dots + c_k\vec{v_k}\), where \(c_1, c_2, \dots, c_k\) are scalars.
    • The span of a collection of vectors is the set of all linear combinations of those vectors.
  • In our discussion on spans, we considered whether a collection of \(m\)-component vectors spanned all of \(\mathbb{R}^m\)

    • For example, does a set of three vectors from \(\mathbb{R}^3\) span all of \(\mathbb{R}^3\)?) and, when not, we tried to determine the geometry of the space spanned by the vectors.
  • It is natural now to start considering whether we can remove any vectors from a collection without changing its span.

Reminders and Today’s Goal

Goals for Today: After today’s discussion, you should be able to…

  • Define the notions of linear independence and linear dependence

  • Test whether a collection \(\left\{\vec{v_1},~\vec{v_2},~\cdots,~\vec{v_p}\right\}\) is linearly independent

  • Explain what happens to the span if you

    • remove a vector from a linearly independent set.
    • remove a vector from a linearly dependent set.

Motivating Linear Independence / Dependence

Consider a collection of non-zero vectors \(V = \left\{\vec{v_1},~\vec{v_2},~\cdots,~\vec{v_p}\right\}\)

  • If we assume that one of the vectors can be written as a linear combination of the remaining vectors in \(V\) (and – for convenience – we assume that we’ve ordered the vectors in \(V\) such that it is \(\vec{v_p}\)), then we have: \(c_1\vec{v_1} + c_2\vec{v_2} + \cdots + c_{p-1}\vec{v_{p-1}} = \vec{v_p}\)
  • Because of the item above, we can write the zero-vector as a linear combination of the vectors in \(V\): \(c_1\vec{v_1} + c_2\vec{v_2} + \cdots + c_{p-1}\vec{v_{p-1}} - \vec{v_p} = \vec{0}\)
  • In fact, there are infinitely many ways to write \(\vec{0}\) as a linear combination of vectors from \(V\) since: \(s\cdot c_1\vec{v_1} + s\cdot c_2\vec{v_2} + \cdots + s\cdot c_{p-1}\vec{v_{p-1}} - s\vec{v_p} = \vec{0}\) for any scalar \(s\)
  • There is some redundancy here.

Motivating Linear Independence / Dependence

Consider a collection of non-zero vectors \(V = \left\{\vec{v_1},~\vec{v_2},~\cdots,~\vec{v_p}\right\}\)

  • If we assume that no vector can be written as a linear combination of the other vectors in \(V\), then the only way to write \(\vec{0}\) as a linear combination of vectors from \(V\) is to have all weights \(c_i = 0\)

    • If there was another way, then we could write \(c_1\vec{v_1} + c_2\vec{v_2} + \cdots + c_p\vec{v_p} = \vec{0}\) where at least one of the \(c_i\) is non-zero.
    • For convenience, we could assume that we’ve ordered the vectors in \(V\) such that \(c_p\), the weight \(\vec{v_p}\), is non-zero.
    • In this case, \(c_1\vec{v_1} + c_2\vec{v_2} + \cdots c_{p-1}\vec{v_{p-1}} = c_p\vec{v_p}\)
    • Which means that \(\frac{c_1}{c_p}\vec{v_1} + \frac{c_2}{c_p} + \cdots + \frac{c_{p-1}}{c_p}\vec{v_{p-1}} = \vec{v_p}\)
    • This contradicts the assumption that \(\vec{v_p}\) is not a linear combination of the other vectors in \(V\).

Definining Linear Independence / Dependence

Definition (Linear Independence): Given a set of vectors \(V = \left\{\vec{v_1},~\vec{v_2},~\cdots,~\vec{v_p}\right\}\), if the only solution to the vector equation \(c_1\vec{v_1} + c_2\vec{v_2} + \cdots + c_p\vec{v_p} = \vec{0}\) is that \(c_1 = c_2 = \cdots = c_p = 0\), then the vectors in \(V\) are linearly independent.

Definition (Linear Dependence): Given a set of vectors \(V = \left\{\vec{v_1},~\vec{v_2},~\cdots,~\vec{v_p}\right\}\), if there exists a solution to the vector equation \(c_1\vec{v_1} + c_2\vec{v_2} + \cdots + c_p\vec{v_p} = \vec{0}\) with at least one \(c_i \neq 0\), then the vectors in \(V\) are linearly dependent.

  • Linear Dependence and Vector Equations: A collection of vectors is linearly dependent if the vector equation \(x_1\vec{v_1} + x_2\vec{v_2} + \cdots + x_p\vec{v_p} = \vec{0}\) has infinitely many solutions.

Notice that reducing a linearly dependent collection of vectors to a linearly independent collection does not change its span.

Definining Linear Independence / Dependence

Definition (Linear Independence): Given a set of vectors \(V = \left\{\vec{v_1},~\vec{v_2},~\cdots,~\vec{v_p}\right\}\), if the only solution to the vector equation \(c_1\vec{v_1} + c_2\vec{v_2} + \cdots + c_p\vec{v_p} = \vec{0}\) is that \(c_1 = c_2 = \cdots = c_p = 0\), then the vectors in \(V\) are linearly independent.

Definition (Linear Dependence): Given a set of vectors \(V = \left\{\vec{v_1},~\vec{v_2},~\cdots,~\vec{v_p}\right\}\), if there exists a solution to the vector equation \(c_1\vec{v_1} + c_2\vec{v_2} + \cdots + c_p\vec{v_p} = \vec{0}\) with at least one \(c_i \neq 0\), then the vectors in \(V\) are linearly dependent.

  • Linear Dependence and Vector Equations: A collection of vectors is linearly dependent if the vector equation \(x_1\vec{v_1} + x_2\vec{v_2} + \cdots + x_p\vec{v_p} = \vec{0}\) has infinitely many solutions.

Notice that reducing a linearly dependent collection of vectors to a linearly independent collection does not change its span. In some sense, we might think of linearly independent collections as being “efficient”.

Homogeneous Systems

Definition (Homogeneous System/Equation): When the right-hand side / constant vector in a system or equation is the zero vector (ie. \(\vec{b} = \vec{0}\)), then that system or equation is called homogeneous

Homogeneous systems and equations are always consistent, because the zero-vector is always a solution.

\[\left\{\begin{array}{lcl} a_{11}x_1 + a_{12}x_2 + \cdots + a_{1p}x_p & = & 0\\ a_{21}x_1 + a_{22}x_2 + \cdots + a_{2p}x_p & = & 0\\ & \vdots & \\ a_{m1}x_1 + a_{m2}x_2 + \cdots + a_{mp}x_p & = & 0 \end{array}\right.\]

Has solution \(x_1 = 0,~x_2 = 0,~\cdots,~x_p = 0\)

Homogeneous Systems

Definition (Homogeneous System/Equation): When the right-hand side / constant vector in a system or equation is the zero vector (ie. \(\vec{b} = \vec{0}\)), then that system or equation is called homogeneous

Homogeneous systems and equations are always consistent, because the zero-vector is always a solution.

\[A\vec{x} = \vec{0} \text{ or } \left[\begin{array}{cccc|c} \vec{a_1} & \vec{a_2} & \cdots & \vec{a_p} & \vec{0}\end{array}\right]\]

Has solution \(\vec{x} = \vec{0}\)

Homogeneous Systems

Definition (Homogeneous System/Equation): When the right-hand side / constant vector in a system or equation is the zero vector (ie. \(\vec{b} = \vec{0}\)), then that system or equation is called homogeneous

Homogeneous systems and equations are always consistent, because the zero-vector is always a solution.

\[x_1\vec{v_1} + x_2\vec{v_2} + \cdots + x_p\vec{v_p} = \vec{0}\]

Has solution \(x_1 = 0,~x_2 = 0,~\cdots,~x_p = 0\)

Homogeneous Systems and Linear Independence

Investigating homogeneous equations provides us with a simple strategy for determining whether a collection of vectors is linearly independent or linearly dependent.

Strategy: We identify whether \(\left\{\vec{v_1},~\vec{v_2},~\cdots,~\vec{v_p}\right\}\) is linearly independent or dependent by determining whether or not the system corresponding to the augmented matrix \(\left[\begin{array}{cccc|c} v_{11} & v_{12} & \cdots & v_{1p} & 0\\ v_{21} & v_{22} & \cdots & v_{2p} & 0\\ \vdots & \vdots & \ddots & \vdots & \vdots\\ v_{m1} & v_{m2} & \cdots & v_{mp} & 0\\ \end{array}\right]\) has a free variable.

Homogeneous Systems and Linear Independence

Strategy: We identify whether \(\left\{\vec{v_1},~\vec{v_2},~\cdots,~\vec{v_p}\right\}\) is linearly independent or dependent by determining whether or not the system corresponding to the augmented matrix \(\left[\begin{array}{cccc|c} v_{11} & v_{12} & \cdots & v_{1p} & 0\\ v_{21} & v_{22} & \cdots & v_{2p} & 0\\ \vdots & \vdots & \ddots & \vdots & \vdots\\ v_{m1} & v_{m2} & \cdots & v_{mp} & 0\\ \end{array}\right]\) has a free variable.

Recall that, since \(\vec{x} = \vec{0}\) is a solution to the homogeneous system, the only way for a non-zero solution to exist is to have infinitely many solutions.

We’ve discovered previously that a system has infinitely many solutions if it has at least one free variable.

Any column of the coefficient matrix which is not a pivot column corresponds to a free variable.

Completed Example #1

Completed Example #2

Linear Independence / Dependence: Special Cases

Although they align with our observations and strategy above, it is worth pointing out a couple of special cases.

  1. Any collection containing a single non-zero vector \(\left\{\vec{v_1}\right\}\) is a linearly independent set.
  2. Any collection including the zero vector \(\vec{0}\) as an element is linearly dependent.

Summary

  • A collection of vectors \(\left\{\vec{v_1},~\vec{v_2},~\cdots,~\vec{v_p}\right\}\) is linearly dependent if at least one of the vectors in the set can be written as a linear combination of the other vectors in the set. Otherwise, the collection is linearly independent.

  • We can use several of the tools/investigations we’ve already encountered to determine whether collections \(\left\{\vec{v_1},~\vec{v_2},~\cdots,~\vec{v_p}\right\}\) are linearly independent.

    • If the corresponding homogeneous system has a unique solution, then the collection is linearly independent, otherwise the collection is linearly dependent.
    • The collection of vectors is linearly independent if the matrix \(\begin{bmatrix} \vec{v_1} & \vec{v_2} & \cdots & \vec{v_p}\end{bmatrix}\) has a pivot in every column. Otherwise, the collection is linearly dependent.

Summary (Cont’d)

Below are a few observations that we can make. Some of them were mentioned earlier in these slides, but others are mentioned here for the first time.

  • A collection containing a single, non-zero vector is linearly independent.

  • Any collection containing the zero vector \(\vec{0}\) is linearly dependent.

  • Any collection of vectors from \(\mathbb{R}^m\) that contains more than \(m\) vectors is linearly dependent.

    • Note. The converse is not necessarily true.

Strategy: The most reliable (broadly applicable) method for determining whether a collection of vectors \(V = \left\{\vec{v_1},~\vec{v_2},~\cdots,~\vec{v_p}\right\}\) is linearly independent is to determine whether the matrix \(\begin{bmatrix} \vec{v_1} & \vec{v_2} & \cdots & \vec{v_p}\end{bmatrix}\) has a pivot in every column.

  • If the answer is yes, then the vectors in \(V\) are linearly independent.
  • If the answer is no, then the vectors in \(V\) are linearly dependent.

Examples to Try #1

Example: Determine whether the collections of vectors below are linearly independent.

  1. \(\left\{\left[\begin{array}{r} 1\\ -2\\ 0\\ 0\\ 3\end{array}\right]\right\}\)

  2. \(\left\{\left[\begin{array}{r} 1\\ 2\end{array}\right], \left[\begin{array}{r} -1\\ 1\end{array}\right], \left[\begin{array}{r} 3\\ 3\end{array}\right]\right\}\)

  3. \(\left\{\left[\begin{array}{r} 1\\ 0\\ 3\end{array}\right], \left[\begin{array}{r} 0\\ 0\\ 0\end{array}\right]\right\}\)

  4. \(\left\{\left[\begin{array}{r} 1\\ 1\\ 0\end{array}\right], \left[\begin{array}{r} 0\\ -2\\ 1\end{array}\right], \left[\begin{array}{r} -1\\ 0\\ 1\end{array}\right]\right\}\)

Examples to Try #2

Example: Determine the value(s) for \(h\) which make \(\vec{v_3} \in \text{span}\left(\left\{\vec{v_1},~\vec{v_2}\right\}\right)\) where \(\vec{v_1} = \left[\begin{array}{r} 1\\ -3\\ 5\end{array}\right]\), \(\vec{v_2} = \left[\begin{array}{r} -3\\ 9\\ 15\end{array}\right]\), and \(\vec{v_3} = \left[\begin{array}{r}2\\ -5\\ h\end{array}\right]\). For which values of \(h\) is \(\left\{\vec{v_1},~\vec{v_2},~\vec{v_3}\right\}\) linearly independent?

Examples to Try #3

Example: Determine the value(s) of \(h\) which make \(\left[\begin{array}{r} 1\\ -2\\ 4\end{array}\right]\), \(\left[\begin{array}{r} -3\\ 7\\ 6\end{array}\right]\), and \(\left[\begin{array}{r} 2\\ 1\\ h\end{array}\right]\) linearly dependent.

Examples to Try #4

Example: What are the possible row echelon forms of the matrices in the following scenarios:

  1. \(A\) is a \(2\times 2\) matrix with linearly dependent columns.
  2. \(A\) is a \(3\times 3\) matrix with linearly independent columns.
  3. \(A\) is a \(4\times 2\) matrix \(\left[\begin{array}{rr} \vec{a_1} & \vec{a_2}\end{array}\right]\) with \(\vec{a_2}\) not a scalar multiple of \(\vec{a_1}\).
  4. \(A\) is a \(4\times 3\) matrix \(\left[\begin{array}{rrr} \vec{a_1} & \vec{a_2} & \vec{a_3}\end{array}\right]\) such that \(\left\{\vec{a_1},~\vec{a_2}\right\}\) is linearly independent and \(\vec{a_3}\) is not in \(\text{span}\left(\left\{\vec{a_1}, \vec{a_2}\right\}\right)\).

Examples to Try #5

Example: How many pivot columns must a \(6\times 4\) matrix have if its columns span \(\mathbb{R}^4\)? Why?

Examples to Try #6

Example: Construct \(3\times 2\) matrices \(A\) and \(B\) such that \(A\vec{x} = \vec{0}\) has a non-trivial solution but \(B\vec{x} = \vec{0}\) has only the trivial solution.

Homework




\[\Huge{\text{Finish Homework 5}}\] \[\Huge{\text{on MyOpenMath}}\]

Next Time…




\(\Huge{\text{Matrix Transformations}}\)