Machine Perception and Cognitive Robotics Laboratory

Neural Cellular Automata

Neural Cellular Automata is basic research into emergence — studying how complex global behavior arises from simple local neural rules on 2D grids. This is original work exploring the frontier of self-organization, not derived from existing Growing Neural CA approaches.

Model Card

2D grid of cells with neural network update rules. Each cell state updated based on local neighborhood and learned parameters. Standard CA topology with differentiable rules.
Target patterns and emergence objectives. Rules learned through backpropagation on the CA dynamics.
Complex emergent patterns from simple learned local rules. Original research direction independent of Growing Neural CA work.
2D only — no 3D implementation yet. Training stability sensitive to learning rate and grid size.

Team

William Edward Hahn, Ph.D.

Founder and Co-Principal Investigator (Co-PI)
Associate Professor

[email protected]

Machine Perception and Cognitive Robotics Laboratory