Differentiable Programming of Reaction-Diffusion Patterns

Alexander Mordvintsev*
Ettore Randazzo*
Eyvind Niklasson*

correspondence to moralex [at] google [dot] com


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.

Presented @ ALife 2021

Supplementary Videos

Figure 2, 3

Figure 4

"lizards" non-uniform diffusion (Fig.4)

"grid_0135" non-uniform diffusion (Fig.4)

"chequered_0121" non-uniform diffusion (Fig.4)

Figure 5

Figure 6



@inproceedings{mordvintsev2021differentiable, title={Differentiable Programming of Reaction-Diffusion Patterns}, author={Mordvintsev, Alexander and Randazzo, Ettore and Niklasson, Eyvind}, booktitle={ALIFE 2021: The 2021 Conference on Artificial Life}, year={2021}, organization={MIT Press} }