Illuminating Diverse Neural Cellular Automata for Level Generation
This work addresses the need for automated and diverse level generation in video game design, offering an incremental improvement by applying quality diversity methods to neural cellular automata.
The authors tackled the problem of generating diverse video game levels by training neural cellular automata (NCA) using a quality diversity approach, specifically CMA-ME, and demonstrated that it produces small NCAs capable of satisfying complex solvability criteria for deterministic agents in games like maze, Sokoban, and Zelda, outperforming a CPPN baseline in exploring level-space.
We present a method of generating diverse collections of neural cellular automata (NCA) to design video game levels. While NCAs have so far only been trained via supervised learning, we present a quality diversity (QD) approach to generating a collection of NCA level generators. By framing the problem as a QD problem, our approach can train diverse level generators, whose output levels vary based on aesthetic or functional criteria. To efficiently generate NCAs, we train generators via Covariance Matrix Adaptation MAP-Elites (CMA-ME), a quality diversity algorithm which specializes in continuous search spaces. We apply our new method to generate level generators for several 2D tile-based games: a maze game, Sokoban, and Zelda. Our results show that CMA-ME can generate small NCAs that are diverse yet capable, often satisfying complex solvability criteria for deterministic agents. We compare against a Compositional Pattern-Producing Network (CPPN) baseline trained to produce diverse collections of generators and show that the NCA representation yields a better exploration of level-space.