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

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

Computation as State Transition

Input

Input is information provided to a system for its rules to read, store, transform, or respond to.

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

What Is Input?

Definition

Input is information provided to a system for its rules to read, store, transform, or respond to.

Input does not change state by itself.

A rule uses input to create state, update state, produce output, or leave the system unchanged.

Input = provided information that a rule can use


Input vs State

Input is information given to the system.

State is the system's current condition during computation.

Example:

Input: 3
Initial state: x = 3
Rule: x -> x + 2
Next state: x = 5

The input helps create the initial state.

The state is what the system continues to update.

Input can start computation, but state is what changes during computation.


Input As Initial Condition

Input can define the starting point of computation.

Example:

Input: 3
Initial state: x = 3

But input and initial state are not always identical.

Example:

Input: user clicks submit
Initial state: form_submitted = true

The input is the event.

The initial state is how the system represents that event.

Input often determines the initial state.


Input Over Time

Input is not always provided once.

One-time input

input -> computation -> output

Continuous input

Input may arrive while computation is already happening.

Example:

Current state: x = 5
Input: +2
Rule: add input to x
Next state: x = 7

Input: +3
Rule: add input to x
Next state: x = 10

Input can influence computation at multiple points.


Input Must Be Recognizable

A system can only process input it is designed to recognize or handle.

Example:

System rule: add numbers

Input: 5     -> valid
Input: "cat" -> invalid

The word "cat" has meaning to a person.

But it is invalid for a system whose rule only adds numbers.

Input must belong to the domain of the system's rules.


Raw Input vs Interpreted Input

The same input can be interpreted differently by different systems.

Example:

Raw input: "5"

One system may treat it as text:

"5" -> count characters -> 1

Another system may treat it as a number:

"5" -> convert to number -> 5
5 -> add 1 -> 6

The raw input is the same.

The system's rules determine how it is handled.

Input is processed according to the representation and rules of the system.


Valid And Invalid Input

Invalid input is not about meaning.

It is about compatibility with the system's rules.

Examples:

System rule: add numbers

Input: 5     -> valid
Input: "cat" -> invalid

"cat" is meaningful, but it is not valid for numeric addition.

5 is valid because it fits the rule.

Invalid input = input that is not compatible with the system's rules.


Input May Be Ignored

Input can be available without affecting a particular step.

Example:

State: x = 5
Input: y = 10
Rule: x -> x + 1

The input y = 10 exists.

But the current rule does not use it.

Next state:

x = 6

The rule only used x.

Input affects computation only when some rule reads or uses it.


Input As Representation

Input is always given in a representable form:

  • numbers
  • symbols
  • text
  • data structures
  • events

For a system to process input, it must be able to distinguish it from other possible inputs.

Example:

Input A: 0
Input B: 1

If the system can distinguish 0 from 1, then different rules can apply.

If the system cannot distinguish them, it cannot treat them differently.

Input must be distinguishable in the form the system receives.


Final Summary

  • Input is provided information that a rule can use.
  • Input does not change state by itself.
  • Input can determine the initial state.
  • Input can arrive once or over time.
  • Input must be recognizable by the system's rules.
  • The same raw input can be interpreted differently by different systems.
  • Invalid input is about rule compatibility, not human meaning.
  • Input may be ignored if no current rule uses it.
  • Input must be represented in a distinguishable form.

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

Continue the series.

Output