Neural Networks can learn any function, even capturing fractal level details.
Approximated the #Mandelbrot set, a classic chaotic fractal, an infinitely complex boundary generated by a simple formula in the complex plane. Every zoom reveals more intricate detail, forever.
In this small-scale project, I trained a neural network to learn the fractal shape from pixel coordinates using PyTorch, NumPy, and Matplotlib. This isn’t curve-fitting. This is pure function learning.
- Added positional encodings to represent spatial patterns
- Increasing epochs and depth for better generalization
- And tuning resolution + loss functions
The model approximated the Mandelbrot set with stunning accuracy, no image inputs, just math.
