reaction&diffusion + scalable moire patterns #webgl

inpainting objective, on a base model. Model tries to learn oscillators to remember the initial picture, but struggles with it

inpainting objective with 1 channel used as hard memory (updates on this channel are zeroed)

https://www.youtube.com/watch?v=t_HcBAO_Yas

угарнейшая история полная самоиронии

5:55 meirl

collider

metastability #webgpu

primordial partical system with high partical speeds and huge interaction radius.

when beta is small negative number, particles tend to cluster together. If beta is a bigger negative number, behaviour becomes chaotic.
As always, there is an interesting region in the parameter space between those two phases

https://compute.toys/view/359

[WIP] refactor NCA to use array structs, leveraging webgpu capabilities #webgpu

simple implementation of texture NCA in webgpu compute shaders
https://compute.toys/view/359
(code for train & export your own model is in the next post)

https://github.com/algroznykh/notebooks/blob/master/%CE%BCNCA_pytorch_compute_codegen.ipynb
code for training own model & export to compute.toys

While refactoring NCA to use webgpu arrays, I've encountered a strange performance problem:
in 1x1 conv stage, when there is 12*48 cycles to compute state update, my machine initially hanged after ~50 cycles.
It turned out, that accessing weights that were defined as top-level const is very inefficient for some reason.
copying weight matrix to a local variable in the function body solved it 🤷

new neural cellular automata demo + recap of recent papers
https://google-research.github.io/self-organising-systems/isonca/

comet