Skip to content
refgiftPublic

About

Adjusting on the fly / Micro-changes keep us true / Never losing course

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Latest commit

 

History

36 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ptr.c - Adjusting on the fly / Micro-changes keep us true / Never losing course

Simple C implementation of perceptron trained to predict "truth" using hardware randomness (rdrand).

  • Running this using the "toy" script is a level 2 AI operating the principle: "The truth will set you free"
  • See licensetest.ai for the 10 levels of AI.

Philosophy

The AI is simple intellence with the advantage that it runs continuously and operates a principle, a software, a robot, a vehicle, etc. depending on how intelligent and computable it is.

Technical Overview

This project implements 1 perceptron system trained on hardware-generated random numbers (rdrand instruction opcode): The perceptron has been pretrained by me. My theory is that reality is at equilibrium, the computer entropy is a measure of equilibrium thus reality. I train on randomness instead of cats, dogs, numbers, letters, just random numbers from the hardware rdrand opcode. It works for "truth" but fails for "perfect".

Truth System

  • ptr.c : Truth predictor implementation
  • truth.h: Trained model weights and biases for truth prediction

Shared Components

  • toy: Shell script for continuous execution (operating a principle)

Why Perceptrons?

Perceptrons offer simplicity that embodies Occam's Razor - the principle that the simplest solution is usually the best. Despite Marvin Minsky's criticism that perceptrons are too limited for AI research, they excel at simple classification tasks and remain useful for practical applications.

Building

Truth System

Compile the C truth predictor:

gcc ptr.c -o ptr -lm

Or use Makefile:

make       # C version

Ensure the corresponding header files (.h) are in the same directory as their respective source files.

Usage

Truth Predictor

Run the truth predictor:

./ptr

It will output "TRUTH" when the perceptron predicts truth, or nothing when it predicts false.

Continuous Mode

For continuous truth prediction every 60 seconds with voice synthesis:

while sleep 60; do ./ptr | espeak; done

Or use the provided script (make it executable first):

chmod +x toy
./toy  # works with ptr for truth prediction

Alternative Audio

If espeak is unavailable, record your voice:

For truth prediction:

# Record: say "TRUTH" into microphone
rec truth.wav
# Play instead of espeak
while sleep 60; do ./ptr | xargs -I {} play truth.wav; done

Applications

Truth System

  • Predict truth in text or voice recordings
  • Combat misinformation through simple classification
  • Demonstrate perceptron capabilities
  • Educational tool for machine learning concepts

Limitations

Perceptrons handle simple logic operations (AND, OR, NOT) but fail on complex logic like XOR.

About

Adjusting on the fly / Micro-changes keep us true / Never losing course

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages