Will Smith
@willsmithvision.bsky.social
Professor in Computer Vision at the University of York, vision/graphics/ML research, Boro @mfc.co.uk fan and climber
📍York, UK
🔗 https://www-users.york.ac.uk/~waps101/
📍York, UK
🔗 https://www-users.york.ac.uk/~waps101/
Our neural network library extends the autograd engine to create neurons, linear layers and MLPs. Constructing an MLP and making a forward pass is as easy as this.
April 1, 2025 at 12:30 PM
Our neural network library extends the autograd engine to create neurons, linear layers and MLPs. Constructing an MLP and making a forward pass is as easy as this.
Just like Karpathy's micrograd, NeuRaLaTeX implements backpropagation (reverse-mode autodiff) over a dynamically built DAG. However, while micrograd uses 150 lines of Python, NeuRaLaTeX uses around 1,100 line of pure LaTeX, making it about 700% better.
April 1, 2025 at 12:30 PM
Just like Karpathy's micrograd, NeuRaLaTeX implements backpropagation (reverse-mode autodiff) over a dynamically built DAG. However, while micrograd uses 150 lines of Python, NeuRaLaTeX uses around 1,100 line of pure LaTeX, making it about 700% better.
So, we wrote a neural net library entirely in LaTeX...
April 1, 2025 at 12:30 PM
So, we wrote a neural net library entirely in LaTeX...