The Timpson Lyceum seal

The Timpson Lyceum · Linear Algebra (2026)

1.5 Exercises

Problems with worked solutions · 20 points · 4 extra credit

Solution Sets of Linear Systems

  1. 1. (2 pts) Solve the homogeneous system and write its solution set in parametric vector form:

    x1−2x2+3x3=02x1−3x2+4x3=0.\begin{aligned} x_1 - 2x_2 + 3x_3 &= 0 \\ 2x_1 - 3x_2 + 4x_3 &= 0. \end{aligned}

    Solution

    1. Reduce the coefficient matrix:

    [1−232−34]→R2−2R1[1−2301−2]→R1+2R2[10−101−2].\begin{bmatrix} 1 & -2 & 3 \\ 2 & -3 & 4 \end{bmatrix} \xrightarrow{R_2 - 2R_1} \begin{bmatrix} 1 & -2 & 3 \\ 0 & 1 & -2 \end{bmatrix} \xrightarrow{R_1 + 2R_2} \begin{bmatrix} 1 & 0 & -1 \\ 0 & 1 & -2 \end{bmatrix}.

    So x1=x3x_1 = x_3, x2=2x3x_2 = 2x_3, with x3x_3 free:

    x=x3[121].\mathbf{x} = x_3 \begin{bmatrix} 1 \\ 2 \\ 1 \end{bmatrix}.

  2. 2. (3 pts) Find the solution set of Ax=0A\mathbf{x} = \mathbf{0} in parametric vector form, where

    A=[12−1324−181205].A = \begin{bmatrix} 1 & 2 & -1 & 3 \\ 2 & 4 & -1 & 8 \\ 1 & 2 & 0 & 5 \end{bmatrix}.

    Solution

    2. Reduce AA:

    [12−1324−181205]→R3−R1R2−2R1[12−1300120012]→R3−R2[12−1300120000]→R1+R2[120500120000].\begin{bmatrix} 1 & 2 & -1 & 3 \\ 2 & 4 & -1 & 8 \\ 1 & 2 & 0 & 5 \end{bmatrix} \xrightarrow[R_3 - R_1]{R_2 - 2R_1} \begin{bmatrix} 1 & 2 & -1 & 3 \\ 0 & 0 & 1 & 2 \\ 0 & 0 & 1 & 2 \end{bmatrix} \xrightarrow{R_3 - R_2} \begin{bmatrix} 1 & 2 & -1 & 3 \\ 0 & 0 & 1 & 2 \\ 0 & 0 & 0 & 0 \end{bmatrix} \xrightarrow{R_1 + R_2} \begin{bmatrix} 1 & 2 & 0 & 5 \\ 0 & 0 & 1 & 2 \\ 0 & 0 & 0 & 0 \end{bmatrix}.

    Pivots in columns 1 and 3, so x2,x4x_2, x_4 are free: x1=−2x2−5x4x_1 = -2x_2 - 5x_4, x3=−2x4x_3 = -2x_4. Thus

    x=x2[−2100]+x4[−50−21].\mathbf{x} = x_2 \begin{bmatrix} -2 \\ 1 \\ 0 \\ 0 \end{bmatrix} + x_4 \begin{bmatrix} -5 \\ 0 \\ -2 \\ 1 \end{bmatrix}.

  3. 3. (3 pts) Solve the system below and express its solution in the form x=p+x2v\mathbf{x} = \mathbf{p} + x_2\mathbf{v}:

    x1+2x2+x3=42x1+4x2+x3=5.\begin{aligned} x_1 + 2x_2 + x_3 &= 4 \\ 2x_1 + 4x_2 + x_3 &= 5. \end{aligned}

    Solution

    3. Augmented reduction:

    [12142415]→R2−2R1[121400−1−3].\left[\begin{array}{ccc|c} 1 & 2 & 1 & 4 \\ 2 & 4 & 1 & 5 \end{array}\right] \xrightarrow{R_2 - 2R_1} \left[\begin{array}{ccc|c} 1 & 2 & 1 & 4 \\ 0 & 0 & -1 & -3 \end{array}\right].

    Row 2 gives x3=3x_3 = 3; then x1=4−2x2−3=1−2x2x_1 = 4 - 2x_2 - 3 = 1 - 2x_2, with x2x_2 free. So

    x=[103]+x2[−210].\mathbf{x} = \begin{bmatrix} 1 \\ 0 \\ 3 \end{bmatrix} + x_2 \begin{bmatrix} -2 \\ 1 \\ 0 \end{bmatrix}.

  4. 4. (3 pts) For the system in Problem 3, write the solution set of the associated homogeneous system Ax=0A\mathbf{x} = \mathbf{0}, and describe geometrically how the two solution sets are related.

    Solution

    4. Dropping the constants (setting the right sides to 00) gives x1=−2x2x_1 = -2x_2, x3=0x_3 = 0, so the homogeneous solution set is { x2[−210] }\left\{\, x_2\begin{bmatrix} -2 \\ 1 \\ 0 \end{bmatrix} \,\right\} — a line through the origin in the direction (−2,1,0)(-2, 1, 0). The solution set of Problem 3 is that same line translated by p=(1,0,3)\mathbf{p} = (1, 0, 3): a parallel line through (1,0,3)(1,0,3) that does not pass through the origin.

  5. 5. (3 pts) State whether each of the following is true or false, and justify each answer.

    (a) A homogeneous system Ax=0A\mathbf{x} = \mathbf{0} is always consistent.

    (b) If Ax=0A\mathbf{x} = \mathbf{0} has only the trivial solution, then Ax=bA\mathbf{x} = \mathbf{b} has a unique solution for every b\mathbf{b}.

    (c) If Ax=bA\mathbf{x} = \mathbf{b} is consistent, then its solution set is a translate of the solution set of Ax=0A\mathbf{x} = \mathbf{0}.

    Solution

    5.

    (a) True. Setting all variables to 00 (the trivial solution) always satisfies Ax=0A\mathbf{x} = \mathbf{0}, so the system is consistent.

    (b) False. "Only the trivial solution" means the columns are independent, so when Ax=bA\mathbf{x}=\mathbf{b} is consistent the solution is unique — but for a non-square AA the equation may be inconsistent for some b\mathbf{b}. Uniqueness for every b\mathbf{b} requires a pivot in every row as well.

    (c) True. This is Theorem 6: for a consistent system with particular solution p\mathbf{p}, the solution set is {p+vh:Avh=0}\{\mathbf{p} + \mathbf{v}_h : A\mathbf{v}_h = \mathbf{0}\}, i.e. a translate by p\mathbf{p} of the solution set of Ax=0A\mathbf{x} = \mathbf{0}.

  6. 6. (3 pts) Describe geometrically, in R3\mathbb{R}^3, the two solution sets, and state how they differ:

    (a) x=t[21−1]\mathbf{x} = t\begin{bmatrix} 2 \\ 1 \\ -1 \end{bmatrix}

    (b) x=[102]+t[21−1]\mathbf{x} = \begin{bmatrix} 1 \\ 0 \\ 2 \end{bmatrix} + t\begin{bmatrix} 2 \\ 1 \\ -1 \end{bmatrix}

    Solution

    6.

    (a) x=t(2,1,−1)\mathbf{x} = t(2,1,-1) traces all scalar multiples of (2,1,−1)(2,1,-1): a line through the origin with direction (2,1,−1)(2,1,-1).

    (b) x=(1,0,2)+t(2,1,−1)\mathbf{x} = (1,0,2) + t(2,1,-1) is the same line translated so it passes through (1,0,2)(1,0,2); it is parallel to the line in (a) but does not contain the origin (since (1,0,2)(1,0,2) is not a multiple of (2,1,−1)(2,1,-1)).

  7. 7. (3 pts) Prove that every homogeneous linear system with more unknowns than equations has a nontrivial solution.

    Solution

    7. Let the system be Ax=0A\mathbf{x} = \mathbf{0} with AA of size m×nm \times n where n>mn > m (more unknowns than equations). The number of pivots is at most m<nm < n, so at least one of the nn columns has no pivot, giving at least one free variable. A homogeneous system is always consistent, and a free variable may be assigned any nonzero value to produce a solution x≠0\mathbf{x} \neq \mathbf{0}. Hence a nontrivial solution exists. ■\blacksquare

  8. 8. (4 pts ••) Suppose Ax=bA\mathbf{x} = \mathbf{b} is consistent, and let p\mathbf{p} be one solution. Prove that the solution set of Ax=bA\mathbf{x} = \mathbf{b} equals {p+vh:Avh=0}\{\mathbf{p} + \mathbf{v}_h : A\mathbf{v}_h = \mathbf{0}\}.

    Solution

    8. Let SS be the solution set of Ax=bA\mathbf{x} = \mathbf{b} and let H={vh:Avh=0}H = \{\mathbf{v}_h : A\mathbf{v}_h = \mathbf{0}\}.

    (S⊆p+HS \subseteq \mathbf{p} + H) Let w∈S\mathbf{w} \in S, so Aw=bA\mathbf{w} = \mathbf{b}. Put vh=w−p\mathbf{v}_h = \mathbf{w} - \mathbf{p}. Then Avh=Aw−Ap=b−b=0A\mathbf{v}_h = A\mathbf{w} - A\mathbf{p} = \mathbf{b} - \mathbf{b} = \mathbf{0}, so vh∈H\mathbf{v}_h \in H and w=p+vh\mathbf{w} = \mathbf{p} + \mathbf{v}_h.

    (p+H⊆S\mathbf{p} + H \subseteq S) Let vh∈H\mathbf{v}_h \in H. Then A(p+vh)=Ap+Avh=b+0=bA(\mathbf{p} + \mathbf{v}_h) = A\mathbf{p} + A\mathbf{v}_h = \mathbf{b} + \mathbf{0} = \mathbf{b}, so p+vh∈S\mathbf{p} + \mathbf{v}_h \in S.

    Both inclusions hold, so S={p+vh:Avh=0}S = \{\mathbf{p} + \mathbf{v}_h : A\mathbf{v}_h = \mathbf{0}\}. ■\blacksquare