I gave the article a brief scan and I was looking for anything NeRF related. I saw the article was implementing an MLP which is the main neural network component of NeRFs.
> The type of neural network that we’re going to use in this article is called multilayer perceptron (MLP). It is relatively simple, but also powerful, and is widely used in practice (e.g., for neural radiance cache (NRC) and neural radiance fields (NeRFs)).
This is the only reference to graphics that I found. The title can be changed a bit to reflect that it’s probably a part 1 from a series.
Edit: after some less brief scanning I found more references to graphics:
> The goal is to train the network to perform a desired mapping between inputs and output values. E.g., the NRC algorithm trains the network to map inputs like surface position and normal to radiance values.
I think the graphics association comes from having written the implementation in HLSL. HLSL compute shaders are general-purpose (like CUDA) but often used in the graphics context, so graphics programmers might feel more comfortable with it?
Tbh I don't really know, I've done extensive ML and graphics work and I'd just pick the right tool for the job (PyTorch and just drop down to Triton/CUDA/whatever when you need custom GPU code).
