Evolving competing neural cellular automata show lifelike self-organizing behavior
burny_tech · x · 2026-10-17
A new paper from @uljadb99's team tackles the famously hard problem of replicating natural evolution's open-endedness in silico. By evolving populations of competing neural cellular automata, they produce lifelike behavior and self-organizing emergent structures, run on the Isambard supercomputer. The thread includes demo videos (with a flashing-lights warning).
More from Research
- Study: AI agents steer wealthier-seeming users toward pricier options across 13 models — niloofar_mire · 2026-10-17
- DSReg post-hoc regularization provably recovers latents without reconstruction; Nature MI paper unifies superposition theory — ninamiolane · 2026-10-17
- VaeZip: open-source codec stores VAE latent datasets 5-10x smaller with near-zero quality loss — TensorForger · 2026-10-17
- GPT-6 Astra nears 100% on Columbia's SRE-Bench reverse-engineering benchmark, others below 60% — banteg · 2026-10-17
- Survive or Collapse: data gating, not rewards, drives self-play RL collapse — shangbinfeng · 2026-10-17
- Goodfire's interpretability breakthrough splits OpenAI researchers' takes — burny_tech · 2026-10-17