https://github.com/daanv2/signal-neural-network
Our neural networks talks with signals, not numbers. They seem to send certain signals in rhythmic patterns. They seem to combine and process these signals and output new signals.
https://github.com/daanv2/signal-neural-network
neural-coding neural-network rust
Last synced: about 1 year ago
JSON representation
Our neural networks talks with signals, not numbers. They seem to send certain signals in rhythmic patterns. They seem to combine and process these signals and output new signals.
- Host: GitHub
- URL: https://github.com/daanv2/signal-neural-network
- Owner: DaanV2
- License: mit
- Created: 2025-07-05T12:52:31.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-07-05T14:02:28.000Z (about 1 year ago)
- Last Synced: 2025-07-05T14:20:44.012Z (about 1 year ago)
- Topics: neural-coding, neural-network, rust
- Language: Rust
- Homepage:
- Size: 6.84 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Signal Neural Network (SNN)
This is based around an idea I have, no idea if it already exists. but I did found the neural coding wiki article.
Our neural networks talks with signals, not numbers.
They seem to send certain signals in rhythmic patterns. They seem to combine and process these signals and output new signals.
```mermaid
xychart-beta
title "Signals Example"
x-axis [0, 0.9, 1, 1.1, 1.9, 2, 2.1, 2.9, 3, 3.1, 3.9, 4, 4.1, 4.9, 5, 5.1, 5.9, 6, 6.1, 6.9, 7, 7.1, 7.9, 8, 8.1, 8.9, 9, 9.1, 9.9, 10, 10.1]
y-axis "" 0 --> 10
line [1, 1, 10, 1, 1, 10, 1, 1, 10, 1, 1, 10, 1, 1, 1, 1, 1, 1, 1, 1, 10, 1, 1, 1, 1, 1, 1]
```
In theory we could simplfy these signals into a binary representation, whereby the peak of the signal is a 1 and the trough is a 0.
so we could represent the above signal as:
```
111100100
```
Now most signals in our brain are more complex than this, since strength of the signal is not the only thing that matters, but also the timing. And we can also represent so much with just a binary representation. This is also just a speculation on how it really works, but I want to try it out and see if it works.
[Read more about neural coding and signal neural network](./docs/snn.md)