CARL paper uses RL to discover and steer self-organizing patterns in Lenia

pyoudeyer · x · 2026-09-10

A new paper introduces CARL, a closed-loop reinforcement learning agent that learns to create self-organizing patterns in the artificial-life simulation Lenia, steer their behavior, and let humans guide them in real time. The team released a project website with interactive demos.

Original post →

More from Research

Research channel →