https://github.com/selfint/beginningoflife
https://github.com/selfint/beginningoflife
Last synced: 8 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/selfint/beginningoflife
- Owner: selfint
- Created: 2020-04-24T16:01:31.000Z (about 6 years ago)
- Default Branch: master
- Last Pushed: 2020-05-01T13:40:33.000Z (about 6 years ago)
- Last Synced: 2025-10-11T23:19:28.978Z (8 months ago)
- Language: C#
- Size: 1.03 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Beginning Of Life
## Goal
Simulation showing the evolution of different life forms in the ocean, simulating the 'beginning
of life', starting from one simple life form. A submarine will be how one can interact with the
world, with the life forms being able to see and interact with it.
## Algorithm
A genetic algorithm will generate the creature's bodies, as well as their 'logic'. Since the point is
to simulate the *beginning* of life, RNN's with a few neurons and connections should be plenty for
any logic these creatures can evolve. Inspired by [NEAT](http://nn.cs.utexas.edu/downloads/papers/stanley.ec02.pdf), the network will evolve in topology as well as in weight and bias values. Neurons
will be blocks, with each neuron connected to all other neurons the creature has.
## Simplifications
Since this an insane goal, many simplifications will be needed in order to get anything significant evolving in a reasonable amount of time.
### Blocks
This is a powerful simplification, as blocks are easy to simulate at scale, and can abstract many
unnecessary complexities of biological shapes (e.g. minecraft).