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

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

Computation as State Transition

Output

Output is the result a system makes available after applying a rule to input, state, or both.

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

What Is Output?

Definition

Output is the result a system makes available after applying a rule to input, state, or both.

Output can be:

  • a returned value
  • a changed state
  • an external effect
  • a signal to another system
  • an intermediate result used by a later step

Output = result made available by computation


Output vs State

State is the system's current condition.

Output is what the system exposes, returns, changes, or makes available from that condition.

Example:

State: x = 5
Rule: return x + 1
Output: 6

The state contains x = 5.

The output is 6.

State is what the system currently has. Output is what the system makes available.


Output As Final Result

In many cases, the final state contains the information used as output.

Example:

Input: 3
Rule: add 2 three times

States:
3 -> 5 -> 7 -> 9

Output: 9

The final state is 9.

The output is the value exposed from that final state.

A final state can determine the final output.


Output As Extracted Information

Output can also be derived from the state.

Example:

State: (x = 5, y = 10)
Rule: return x + y
Output: 15

The output is not the whole state.

It is extracted from part of the state.

Output may be a value derived from the current state.


Output As Effect

Output is not always a returned value.

It can also be an effect.

Examples:

print("hello") -> text appears on screen
delete file -> file no longer exists
send signal -> another system receives it

In these cases, the output is what the computation causes or exposes.

Output can be a value, a state change, or an observable effect.


Output Does Not Always Change State

Output may leave the internal state unchanged while still creating an external effect.

Example:

Internal state: x = 5
Rule: print x
External output: 5
Internal state after output: x = 5

The internal state did not change.

But something was still produced outside the internal state.

Output can expose information without changing the internal state.


Output Over Time

Output is not always a single result.

Single output

input -> computation -> final output

Continuous output

Output may be produced during computation.

Example:

State: 3 -> output: 3
State: 5 -> output: 5
State: 7 -> output: 7
State: 9 -> output: 9

Each output is produced before the full sequence is finished.

Output can be generated at intermediate steps.


Output Must Be Available

Output must be available to some observer, system, or later step.

Examples:

  • displayed value
  • returned result
  • saved file
  • signal to another system
  • value passed into the next step

If nothing can access or use the result, then it does not function as output for that system.

Output must be representable and accessible.


No Final Output

Some computations may never produce a final output.

Example:

while true:
  keep running

The system may keep changing state.

It may even produce intermediate outputs.

But if it never stops, it does not produce a final result.

A computation can have no final output if it never reaches a final result.


Output As Input

Output can feed into further computation.

Example:

First computation:
Input: 3
Rule: add 2
Output: 5

Next computation:
Input: 5
Rule: add 2
Output: 7

What counts as output in one step may become input for another step.

Output of one computation can become input to another.


Raw Output vs Interpreted Output

The same output can be interpreted differently by different systems.

Example:

Output: "7"

One system may treat it as text:

"7" -> count characters -> 1

Another system may treat it as a number:

"7" -> convert to number -> 7
7 -> add 1 -> 8

The raw output is the same.

The interpretation depends on the system that receives it.

Output becomes useful only relative to a system that can handle it.


Output vs Meaning

Output does not require meaning.

Example:

Output: "xk3#9!"

This may have no human meaning.

But it can still be output if it is produced by the system and made available.

Output is a structured result, not necessarily meaningful interpretation.


Final Summary

  • Output is a result made available by computation.
  • Output may come from input, state, or both.
  • State is what the system currently has; output is what it exposes, returns, changes, or makes available.
  • A final state can determine the final output.
  • Output can be extracted from part of the state.
  • Output can be a returned value, state change, signal, or external effect.
  • Output can expose information without changing internal state.
  • Output may be produced once, continuously, or at intermediate steps.
  • Some computations may never produce a final output.
  • Output of one computation can become input to another.
  • Output does not require human meaning.

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

Continue the series.

Memory