Functions

What Is a Function?

A function is a rule that tells us how to convert an input into an output.

A simple way to think about it is:

Input
  ↓
Function
  ↓
Output

For example, imagine a function that doubles a number.

Input = 5

Function = multiply by 2

Output = 10

So the function takes 5 and produces 10.

A Simple Mathematical Function

We can write the same idea mathematically.

f(x) = 2x

Here:

• f represents the function.
• x represents the input.
• 2x tells us what calculation to perform.

If the input is 5:

f(5) = 2 × 5

     = 10

Therefore, the output is 10.

Think of a Function Like a Machine

One of the easiest ways to understand a function is to imagine a machine.

              FUNCTION MACHINE

Input ──────→ [ × 2 ] ──────→ Output

  5                           10

The machine has one job: multiply the input by 2.

So:

Input    Calculation       Output

  2      2 × 2               4
  5      5 × 2              10
 10     10 × 2              20

The same function can be used with different inputs.

Functions in Python

Python also allows us to create functions.

def double(x):
    return x * 2

print(double(5))

Output:

10

Here:

• double is the function name.
• x is the input.
• x * 2 is the calculation.
• return gives us the output.

Another Simple Example

Let's create a function that calculates the square of a number.

def square(x):
    return x * x

print(square(3))
print(square(5))
print(square(10))

Output:

9
25
100

Mathematically, this function can be written as:

f(x) = x²

For example:

f(5)

= 5²

= 25

Function Input and Output

A function can have different inputs and produce different outputs.

Function:

f(x) = x + 10

If the input is 5:

f(5) = 5 + 10
     = 15

If the input is 20:

f(20) = 20 + 10
      = 30

The rule stays the same. Only the input changes.

Why Are Functions Important in AI?

AI models receive information as input and perform mathematical operations to produce an output.

Input Data
    ↓
Mathematical Function
    ↓
Calculation
    ↓
Output / Prediction

For example, suppose an AI model receives information about a house:

House size = 1500 sq ft
Bedrooms   = 3
Location   = 8

These values can be given to mathematical functions inside the model.

House Data
    ↓
Functions
    ↓
Mathematical Calculations
    ↓
Prediction

Predicted Price = ₹75,00,000

A real AI model is much more complicated than this example, but the basic idea is the same: inputs are transformed into outputs through mathematical operations.

Functions Can Be Combined

AI models usually don't use only one function. Multiple calculations can be performed one after another.

Input
  ↓
Function 1
  ↓
Function 2
  ↓
Function 3
  ↓
Output

For example:

Input
  ↓
Multiply by 2
  ↓
Add 5
  ↓
Output

If the input is 10:

10 × 2 = 20

20 + 5 = 25

Final Output = 25

Neural networks perform many transformations like this, although the actual calculations involve vectors, matrices, weights, activation functions, and other mathematical operations.

Functions and Machine Learning

A machine learning model can be viewed as a mathematical function that learns useful parameters from data.

Input Data
    ↓
Machine Learning Model
    ↓
Prediction

For example:

Study Hours
     ↓
ML Model
     ↓
Predicted Exam Score

The model learns how the input values relate to the output values.

This is why understanding functions is important before learning derivatives, gradients, and gradient descent.

Function → Derivative → Gradient Descent

Functions are the starting point for the calculus concepts that come next.

Function
   ↓
How does the output change?
   ↓
Derivative
   ↓
How do many parameters affect the output?
   ↓
Gradient
   ↓
Which direction should we move?
   ↓
Gradient Descent
   ↓
Improve the AI model

So functions are not an isolated mathematical topic. They provide the foundation for understanding how AI models calculate and learn.