~/tedtes/writing/math-foundations-for-reasoning-models

~/writing/math-foundations-for-reasoning-models $ ls

Calculus and Gradients

Function as Input-Output Machine

A function maps input values to output values, giving calculus a precise object whose change can be studied.

~/writing/math-foundations-for-reasoning-models $ cat topic.md

Function As Input-Output Machine

Definition

A function is a rule that maps an input to an output.

Basic form:

input -> function -> output

Example:

f(x) = x + 2

f(3) = 5
f(10) = 12

Function = rule-based input-output mapping.


Why Start With Functions?

Calculus studies how outputs change when inputs change.

To talk about change clearly, we need something that connects input and output.

That object is a function.

Example:

x -> x^2

If the input changes from 2 to 3, the output changes from 4 to 9.

Calculus needs functions because change must be change of something with respect to something else.


Function vs Value

A value is one result.

A function is the rule that can produce many results.

Example:

f(x) = x + 1

Values produced by the function:

f(1) = 2
f(2) = 3
f(3) = 4

The function is not just 2, 3, or 4.

It is the rule connecting inputs to outputs.

A value is one point; a function describes the pattern across points.


Function As Transformation

Earlier, a function was introduced as a transformation.

Example:

x -> x + 1

The input is transformed into another value.

In calculus, we ask a more precise question:

How does the output change as the input changes?

Calculus studies the behavior of transformations under change.


Function With Parameters

Some functions contain parameters.

Example:

f(x) = wx + b

Here:

x = input
w = weight
b = bias

Changing x changes the output.

Changing w or b also changes the output.

Parameters are adjustable values inside a function.


Why Parameters Matter

In the perceptron chapter, the model used:

score = w1*x1 + w2*x2 + b

That is a function.

The output depends on:

input values
weights
bias

Learning later becomes the problem of changing parameters so the function behaves better.

A model is a function with adjustable parameters.


Function Does Not Require Meaning

A function does not need to understand the input.

Example:

f(x) = x^2

The function does not know what x means.

It only applies a rule.

This continues the earlier idea:

computation operates on structure, not awareness

A function can transform values without understanding them.


Final Summary

  • A function maps inputs to outputs.
  • A value is one result; a function is the rule producing results.
  • Calculus studies how function outputs change.
  • Parameters are adjustable values inside a function.
  • Models can be understood as functions with parameters.
  • This prepares the ground for limits, derivatives, and gradients.

~/writing/math-foundations-for-reasoning-models $ open

Continue the series.

Limit