Show HN: Neural Particle Automata

Category: ai-ml

Tags: neural-cellular-automata, particle-system, self-organization, cuda, jupyter-notebook

Score: 7.3/10 (Innovation: 8, Technical: 9, Documentation: 7, Utility: 5)

Neural Particle Automata introduces a novel approach to self-organizing particle dynamics by applying neural cellular automata principles to continuous, irregular particle systems instead of fixed grids. It combines SPH perception with learned update rules, enabling morphogenesis and texture generation, with interactive web demos and Colab notebook access. Its acceptance at SIGGRAPH 2026 and use of custom CUDA kernels highlight its technical depth and potential for generative simulations.

Target audience: researchers, ai engineers, simulation developers

Repository: https://selforg-npa.github.io/ · Jupyter Notebook · 1 stars

View on Hacker News