Linear Equations and Their Solutions

A linear equation draws a straight line, and solving it finds where that line crosses a target. Trend lines, break-even points and single-neuron models all live here.

The line 4x minus 1 climbs through the point (3, 11), so solving 4x minus 1 equals 11 lands exactly on x equals 3 where the line meets its target

Why Does This Exist?

Straight lines are the atoms of ML models. Linear regression fits one. A single neuron computes one before its activation. A decision boundary in logistic regression is one. Each of these is a linear equation, and solving it means finding where the line meets a target: what input makes the output exactly 11, where cost equals revenue, which score sits on the boundary.

The geometry matters as much as the algebra. A line has a slope (how fast it climbs) and an intercept (where it starts), and those two numbers are the entire model. Pairs of lines, and what their crossings mean, graduate to systems of linear equations; fitting lines through noisy data graduates to linear systems and least squares.

The one idea you need first. An equation balances two sides, and isolating $x$ by undoing operations is the move. If that feels new, algebra teaches the loop before this page uses it on lines.

Think of It Like This

A taxi meter and a fare target

A taxi charges a $3 flagfall plus $4 per mile: the fare for $x$ miles is $f = 4x + 3$ . Handed a $15 bill and asked how far it goes, you solve $4x + 3 = 15$ : undo the flagfall, undo the rate, $x = 3$ miles.

The slope is the per-mile rate, the intercept is the flagfall. Every linear equation is a meter with different pricing, and solving is asking how much ride a fixed bill buys.

Where it stops: meters never run backwards into negative miles, while lines extend forever in both directions. The analogy covers the solving; the infinite extension is pure math, and models exploit it.

How It Actually Works

A linear equation in one variable has the form $ax + b = 0$ with $a \neq 0$ : no squares, no roots, $x$ to the first power only. Solve by isolation: $2x + 3 = 11$ gives $2x = 8$ , so $x = 4$ .

The line form is $y = mx + b$ . The slope $m$ is rise over run: through points $(x_1, y_1)$ and $(x_2, y_2)$ , $m = (y_2 - y_1)/(x_2 - x_1)$ . The intercept $b$ is where the line crosses $y$ at $x = 0$ .

Three fates exist. One solution when $a \neq 0$ . No solution when the equation collapses to a falsehood: $0x = 5$ is never true. Infinitely many when it collapses to a truth: $0x = 0$ holds for every $x$ . Code that divides by $a$ without checking courts the second fate.

Worked example: a solution and a line through two points

Solve $2x + 3 = 11$ : subtract 3 to get $2x = 8$ , divide by 2 to get $x = 4$ . Check: $2 \times 4 + 3 = 11$ . Balances.

Fit a line through $(1, 3)$ and $(3, 11)$ . Slope: $m = (11 - 3)/(3 - 1) = 8/2 = 4$ . Using $(1, 3)$ : $y - 3 = 4(x - 1)$ , so $y = 4x - 1$ . Check with the second point: $4 \times 3 - 1 = 11$ . A third check at $x = 2$ : $y = 7$ , the midpoint of the two points, exactly where a straight line should pass.

Code

1def solve_linear_1d(a: float, b: float, target: float) -> float:2    # Solve a*x + b = target; caller guarantees a is nonzero.3    return (target - b) / a4
5
6print(solve_linear_1d(2.0, 3.0, 11.0))  # -> 4.07
8x1: float = 1.09y1: float = 3.010x2: float = 3.011y2: float = 11.012slope: float = (y2 - y1) / (x2 - x1)13print(slope)  # -> 4.0

The guard is a comment because the function cannot check intent: a zero slope means no solution or infinitely many, and only the caller knows which. The slope block is rise over run transcribed directly.

Watch Out For

Dividing by a slope that is zero

x = (target - b) / a with $a = 0$ raises in Python and returns inf in NumPy, and both are wrong answers to the real question. $a = 0$ means the line is flat: either the target is unreachable (no solution) or every $x$ works (infinitely many). The code answers a different question than the one asked.

Branch on the slope first: if it is zero, compare the target against the constant directly. Never let a division discover horizontality for you.

Assuming two lines always cross

Parallel lines share a slope and never meet: $y = 2x + 1$ and $y = 2x + 5$ have no common point, so solving them together yields $1 = 5$ , a contradiction. Fitting code that assumes a crossing returns whatever the algebra produces, usually garbage with full confidence.

Compare slopes before solving pairs. Equal slopes mean no crossing (or total overlap), and that needs a decision, not a division.

The Quick Version

Linear means $x$ to the first power only; the graph is always a straight line.
$y = mx + b$ : slope is rise over run, intercept is the value at $x = 0$ .
$2x + 3 = 11$ gives $x = 4$ ; through $(1,3)$ and $(3,11)$ runs $y = 4x - 1$ .
Zero slope means no solution or infinitely many: branch, do not divide.
Parallel lines never cross; check slopes before solving pairs.