Lesson 14.4 — Functions & Graphs

1 Introduction: How Input Translates to Output

Imagine you operate an online e-commerce platform where a product sells for a flat price of $500 per unit. The total amount charged to a customer depends completely on the number of items ordered.

Mathematical Model: y = 500x

x = Input (Units), y = Output (Dollars)

Order Quantity (Input x)	Computation (500 × x)	Total Payable (Output y)	Coordinate Pair (x, y)
1	500 × 1	$500	(1, 500)
2	500 × 2	$1,000	(2, 1000)
3	500 × 3	$1,500	(3, 1500)
4	500 × 4	$2,000	(4, 2000)
5	500 × 5	$2,500	(5, 2500)

Notice that for every valid input, there is exactly one guaranteed output. In mathematics, this deterministic, predictable transformation rule is known as a Function. Whether calculating server network throughput, rendering vector graphics in games, or training artificial neural networks, functions are the universal building blocks of computation.

2 What is a Function?

A Function is a mathematical relationship where each valid element of an input set maps to exactly one distinct element of an output set.

The Function Machine Model

Input (x) x = 2

Function Rule f(x)

f(x) = 2x + 3

Multiply input by 2, then add 3

Output f(2) f(2) = 7

Inviolable Axiom of Functions:

A single input CANNOT produce multiple conflicting outputs. However, multiple distinct inputs CAN map to the exact same output.

✓ Valid Function: f(x) = x² f(2) = 2² = 4
f(−2) = (−2)² = 4
(Different inputs 2 and −2 produce identical output 4 — fully valid!)

✗ NOT a Function: y = ±√x If x = 4, then y = 2 and y = −2
(A single input 4 yields two distinct outputs — violates function definition!)

3 Function Notation

The universal shorthand for functions is f(x) (pronounced "f of x"). Here f designates the name of the function rule and x represents the input argument.

Two Equivalent Representations:

f(x) = 3x + 2 (Functional Notation)

y = 3x + 2 (Coordinate Equation Form)

When plotting curves, y and f(x) denote identical values: y = f(x). Note: f(x) does NOT mean f × x!

Function Evaluation:

For function f(x) = 3x + 2, evaluate f(4):

f(4) = 3(4) + 2

f(4) = 12 + 2

f(4) = 14

Input 4 produces output 14, establishing the geometric coordinate pair (4, 14).

4 Domain, Codomain, and Range

In computational mathematics, defining the allowable scope of inputs and outputs is vital for avoiding runtime exceptions and ensuring model validity:

4.1

Domain

The complete set of all legal input values for which the function is mathematically defined.

f(x) = 1 / x

x ≠ 0 (division by 0 undefined)

Domain = ℝ \ {0}

4.2

Codomain

The target set containing all possible designated values that outputs could potentially take.

f: ℝ → ℝ

Codomain = All Real Numbers (ℝ)

4.3

Range

The exact set of actual output values generated by evaluating the function across its domain.

f(x) = x² (where x ∈ ℝ)

Range = [0, ∞)

(squares are never negative)

Canonical Relationship: The Range is always a subset of the Codomain (Range ⊆ Codomain). In database schema terminology, the Codomain is the declared column type (e.g. `float64`), while the Range represents the actual bounds of stored data.

5 The Cartesian Coordinate Plane

Visualizing continuous functional behavior requires René Descartes' 2D Cartesian Coordinate Plane, constructed by intersecting two perpendicular number lines:

The two perpendicular axes establish an address system:

X-axis (Horizontal Abscissa): Represents the independent input values.
Y-axis (Vertical Ordinate): Represents the dependent output values.
Origin: The central intersection point with coordinates (0, 0).

Every point is an ordered pair (x, y). For instance, (2, 3) denotes 2 units right and 3 units up from the origin.

Quadrant	x Sign	y Sign	Sample Point
Quadrant I	+ (Right)	+ (Up)	(3, 4)
Quadrant II	− (Left)	+ (Up)	(-3, 4)
Quadrant III	− (Left)	− (Down)	(-3, -4)
Quadrant IV	+ (Right)	− (Down)	(3, -4)

6 What is a Function's Graph?

The Graph of a function is the geometric locus formed by plotting all input-output ordered pairs (x, f(x)) on the Cartesian plane and joining them into a continuous curve.

Constructing the Graph of y = 2x + 1

Linear Graph

Plotting discrete coordinate points:

x = −2 → y = 2(−2)+1 = −3 → (−2, −3)

x = −1 → y = 2(−1)+1 = −1 → (−1, −1)

x = 0 → y = 2(0)+1 = 1 → (0, 1)

x = 1 → y = 2(1)+1 = 3 → (1, 3)

x = 2 → y = 2(2)+1 = 5 → (2, 5)

Key Insights from the Line:

Connecting these points produces an unbroken straight line.
Every 1 unit increase in x causes a constant 2 unit increase in y (Slope = 2).
The line intercepts the Y-axis at (0, 1) (Y-intercept = 1).

7 Taxonomy of Functions

Depending on algebraic structure and polynomial degree, functions exhibit distinctive geometric signatures:

7.1 Linear Function

Straight Line

f(x) = mx + c

Constant rate of change. Parameter m is the slope and c is the Y-intercept.

f(x) = 2x + 1

7.2 Quadratic Function

Parabola

f(x) = ax² + bx + c

Degree 2 polynomial generating a smooth U-shaped curve called a Parabola with a distinct vertex.

f(x) = x² − 4

7.3 Constant Function

Horizontal Line

f(x) = c

Output is entirely invariant regardless of input variations. Produces a flat horizontal line with slope 0.

f(x) = 5

8 The Vertical Line Test

The Vertical Line Test is an infallible visual test to determine whether a given geometric curve qualifies as a valid function.

Execution Principle:

Imagine sweeping a vertical straight line across the entire domain of a graph:

Valid Function (Passed) If every conceivable vertical line intersects the curve at at most one single point, the graph represents a valid function.
Examples: y = 2x + 1, y = x², y = x³

Not a Function (Failed) If any vertical line intersects the curve at two or more points, the relation is NOT a function (violates the single-output rule).
Examples: Circles (x² + y² = r²), sideways parabolas

9 Real-World & Computational Applications

Functions power modern software architectures, data science regressions, and machine learning neural models.

9.1 Cloud Infrastructure Cost Curves

Fixed cluster overhead of $2,000 plus $100 per container:

C(x) = 2000 + 100x (Linear Cost Curve)

9.2 Data Trend Analysis & Regressions

Fitting mathematical curves across empirical scatter plots to identify customer churn and forecast revenue.

9.3 Python Functions: Pure Functions vs. Side Effects

# Pure mathematical function in Python
def calculate_order_total(quantity: int) -> float:
    unit_price = 500
    return quantity * unit_price  # Deterministic mapping without side effects

print(calculate_order_total(4))  # 2000

Like mathematical functions, a software routine that consistently returns identical results for identical arguments without mutating external state is called a Pure Function.

9.4 Activation Functions in AI: Non-Linearity

Without non-linear transformations, deep neural networks would collapse into simple linear regressions. Special non-linear functions called Activation Functions allow models to learn complex visual and linguistic representations:

ReLU (Rectified Linear Unit): f(x) = max(0, x)
Zeroes negative signals; passes positive signals directly.

Sigmoid Function: σ(x) = 1 / (1 + e^−x)
Compresses any real number into a probability [0, 1].

10 5 Common Pitfalls in Functions & Graphs

Avoid these classic conceptual misconceptions:

Pitfall 1: Treating f(x) as Multiplication

f(x) ≠ f × x

f is the function name; x is the input argument.

Pitfall 2: Assuming All Relations are Functions

x² + y² = 25

A circle is not a function because single inputs yield two outputs.

Pitfall 3: Conflating Domain with Range

Domain = allowable inputs; Range = actual outputs produced.

They serve entirely different mathematical roles.

Pitfall 4: Inverting Coordinate Order

(2, 3) ≠ (3, 2)

The first value is always X; the second is always Y.

Hands-on Interactive Workbench

Functions & Graphs Laboratory

Simulate the function machine, test relations with the vertical line sweep, and experiment with AI activation curves.

Function Selection & Live Plotting

Select a function family, adjust input x, and watch the point trace along the coordinate curve in real time.

Function:

Blue Line: Function Curve | Red Dot: Current State (x, f(x))

Input Argument (x): 1.0

Evaluation Pipeline:

f(1.0) = 2(1.0) + 1

Output y = 3.0

Coordinates: (1.0, 3.0)

Lesson Summary (Key Takeaways)

✓ Function: A deterministic mathematical rule relating every valid input to exactly one output value.
✓ Domain and Range: The Domain is the complete set of legitimate inputs; the Range is the actual set of generated outputs.
✓ Coordinate Plane & Graph: Plotting ordered pairs (x, y) on the Cartesian plane transforms algebraic expressions into visual geometric curves.
✓ Function Diversity: Linear functions produce straight lines, Quadratic functions yield parabolas, and Constant functions form horizontal lines.
✓ Vertical Line Test: If any vertical line intersects a graph in more than one point, the graph cannot represent a function.
✓ AI Foundation: Deep neural networks are composite chains of functions; activation functions like ReLU and Sigmoid introduce the non-linearity necessary for complex pattern recognition.