Identity Increases Stability in Neural Cellular Automata

2025-08-08

AI Summary

Neural Cellular Automata (NCAs) commonly exhibit morphological instabilities in 2D artificial organism growth, including boundary collapse, tumor-like aberrant proliferation, and failure to maintain stable shape. To address this, we propose a lightweight identity layer mechanism that imposes a single scalar identity constraint within the NCA state space—requiring no additional network parameters or complex supervision. This approach markedly enhances structural integrity: in dense multi-organism configurations, individuals retain sharp boundaries and intrinsic morphology, with tumor-like growth reduced by over 70%. Remarkably, without explicit interaction design, stable, coordinated collective motion spontaneously emerges among distinct individuals—an observation unprecedented in prior NCA-based artificial life systems. Our work establishes a new paradigm for scalable, interpretable artificial life modeling and enables principled investigation of cell-level sociality.

Technology Category

• Intelligent Robots: Embodied AI
• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Neural Cellular Automata (NCAs) offer a way to study the growth of two-dimensional artificial organisms from a single seed cell. From the outset, NCA-grown organisms have had issues with stability, their natural boundary often breaking down and exhibiting tumour-like growth or failing to maintain the expected shape. In this paper, we present a method for improving the stability of NCA-grown organisms by introducing an 'identity' layer with simple constraints during training. Results show that NCAs grown in close proximity are more stable compared with the original NCA model. Moreover, only a single identity value is required to achieve this increase in stability. We observe emergent movement from the stable organisms, with increasing prevalence for models with multiple identity values. This work lays the foundation for further study of the interaction between NCA-grown organisms, paving the way for studying social interaction at a cellular level in artificial organisms.

Problem

• Improving stability in Neural Cellular Automata grown organisms
• Preventing boundary breakdown and tumor-like growth patterns
• Introducing identity layer constraints during training for stability

Innovation

• Introducing identity layer with constraints
• Single identity value increases stability
• Multiple identity values enable emergent movement