Lifelike behaviors like colonization and macro-structures can emerge in a digital petri dish without any biological programming.
April 14, 2026
Original Paper
Evolving Many Worlds: Towards Open-Ended Discovery in Petri Dish NCA via Population-Based Training
arXiv · 2604.11248
The Takeaway
A meta-evolutionary algorithm (PBT-NCA) generated spontaneous survival strategies from simple local cellular interactions. It proves that complex, biological-like evolution can be fully simulated and discovered in purely artificial environments.
From the 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 noise. In this paper, we introduce PBT-NCA, a meta-