Evolving Many Worlds: Towards Open-Ended Discovery in Petri Dish NCA via Population-Based Training

arXiv:2604.11248v1 Announce Type: cross Abstract: The generation of sustained, open-ended complexity from local interactions remains a fundamental challenge in artificial life. Differentiable multi-agent systems, such as Petri Dish Neural Cellular Automata (PD-NCA), exhibit rich self-organization driven purely by spatial competition; however, they are highly sensitive to hyperparameters and frequently collapse into uninteresting patterns and dynamics, such as frozen equilibria or structureless n