Differentiable Programming of Reaction-Diffusion Patterns
This addresses the trial-and-error parameter search issue in computational pattern formation for researchers in computer graphics or simulation.
The authors tackled the problem of designing Reaction-Diffusion systems for pattern formation by proposing a differentiable optimization method to learn parameters from examples, achieving robust, non-trivial 'life-like' behavior in synthesized textures.
Reaction-Diffusion (RD) systems provide a computational framework that governs many pattern formation processes in nature. Current RD system design practices boil down to trial-and-error parameter search. We propose a differentiable optimization method for learning the RD system parameters to perform example-based texture synthesis on a 2D plane. We do this by representing the RD system as a variant of Neural Cellular Automata and using task-specific differentiable loss functions. RD systems generated by our method exhibit robust, non-trivial 'life-like' behavior.