Evolving Many Worlds: Open-Ended Discovery with Cellular Automata

Created on April 20, 2026 by Uljad Berdica

2026   ·   neural-cellular-automata   open-endedness   meta-optimization   novelty-search   research   ·   AI   Research   Complexity   Emergent

If you want the full details, visuals, and links to the preprint and code, see the official project page here: PBT-NCA · Evolving Many Worlds.

This blog post highlights open-ended discovery in Neural Cellular Automata via novelty-driven population-based training.



Enjoy Reading This Article?

Here are some more articles you might like to read next:

  • Intent Factored Generation: Best way to use your LLM
  • Shkodra Elektronike - A way to remember
  • The Rise of the Performative Audience