Skip to content

Life simulation that demonstrates neural networks based on the Perceptron algorithm.

License

Notifications You must be signed in to change notification settings

reicek/AI-world

Repository files navigation

Build Status GitHub version Maintainability issues license Donate

AI World

Javascript Neural Network, Perceptron architecture.

by César Antón Dorantes @reicek

Life simulation that demonstrates neural networks based on the perceptron algorithm adjusting to a dynamic scenario.

Creatures continously learn how to move as a group while adjusting to changes.

When fully developed, creatures reproduce when close. After reproduction, both parents lose half their mass and have an offspring of a quarter of the initial parents mass. Chances of breeding increase or decrease in response to the number of living creatures.

Creatures age on each cycle, aging translates into a loss of maximum possible speed. Creatures die when they lose all mobility.

The simulation attempts to keep creatures alive for as long as possible, without reaching overall overpopulation or full extinction.

Install

Run from console:

npm install

Development

Run from console:

npm start

Open localhost:8080 on a browser tab.

Install browser extension to use Live Reload.


Inpired on the demo available at the Synaptic's home page.

About

Life simulation that demonstrates neural networks based on the Perceptron algorithm.

Resources

License

Stars

Watchers

Forks

Packages

No packages published