mlsanityData to 7 Oct 2026

Illuminating Diverse Neural Cellular Automata for Level Generation

arXiv 2109.05489 ↗First posted 12 Sep 2021cs.NE cs.AIcited by 35

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 35 papers in the map (data through 7 Oct 2026). Builds on 28 of its 48 references that are in the map. First posted on arXiv 12 Sep 2021.

Builds on

28 of 48 references.

References · 28

  1. 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,250
  2. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,214
  3. 2015Illuminating search spaces by mapping elitescited by 406
  4. 2016The CMA Evolution Strategy: A Tutorialcited by 364
  5. 2016FractalNet: Ultra-Deep Neural Networks without Residualscited by 216
  6. 2014Robots that can adapt like animalscited by 203
  7. 2021Open-Ended Learning Leads to Generally Capable Agentscited by 177
  8. 2017Procedural Content Generation via Machine Learning (PCGML)cited by 136
  9. 2017Improving Exploration in Evolution Strategies for Deep Reinforcement Learning via a Population of Novelty-Seeking Agentscited by 107
  10. 2020PCGRL: Procedural Content Generation via Reinforcement Learningcited by 90
  11. 2019Covariance Matrix Adaptation for the Rapid Illumination of Behavior Spacecited by 78
  12. 2021Differentiable Quality Diversitycited by 71
16 more
  1. 2020Quality-Diversity Optimization: a novel branch of stochastic optimizationcited by 59
  2. 2020Illuminating Mario Scenes in the Latent Space of a Generative Adversarial Networkcited by 51
  3. 2021Learning Controllable Content Generatorscited by 40
  4. 2020Scaling MAP-Elites to Deep Neuroevolutioncited by 38
  5. 2021Growing 3D Artefacts and Functional Machines with Neural Cellular Automatacited by 38
  6. 2020Lenia and Expanded Universecited by 23
  7. 2019Empowering Quality Diversity in Dungeon Design with Interactive Constrained MAP-Elitescited by 18
  8. 2020Video Game Level Repair via Mixed Integer Linear Programmingcited by 18
  9. 2020BOP-Elites, a Bayesian Optimisation algorithm for Quality-Diversity searchcited by 15
  10. 2019Intrinsically Motivated Discovery of Diverse Patterns in Self-Organizing Systemscited by 13
  11. 2020Mixed-Initiative Level Design with RL Brushcited by 13
  12. 2020It's Hard for Neural Networks To Learn the Game of Lifecited by 13
  13. 2018Are Efficient Deep Representations Learnable?cited by 6
  14. 2019Evolving Structures in Complex Systemscited by 6
  15. 2020Using Fractal Neural Networks to Play SimCity 1 and Conway's Game of Life at Variable Scalescited by 6
  16. 2021Self-Referential Quality Diversity Through Differential Map-Elitescited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 14; 2023: 11; 2024: 5; 2025: 2; 2026: 3.

35 papers.

Sort:

By area:

  1. 2022Evolving Curricula with Regret-Based Environment Designcited by 166
  2. 2022Deep Surrogate Assisted Generation of Environmentscited by 40
  3. 2023pyribs: A Bare-Bones Python Library for Quality Diversity Optimizationcited by 29
  4. 2022Covariance Matrix Adaptation MAP-Annealingcited by 24
  5. 2023Proximal Policy Gradient Arborescence for Quality Diversity Reinforcement Learningcited by 20
  6. 2022Learning Controllable 3D Level Generatorscited by 19
  7. 2023Stabilizing Unsupervised Environment Design with a Learned Adversarycited by 17
  8. 2024DreamCraft: Text-Guided Generation of Functional 3D Environments in Minecraftcited by 17
  9. 2022Accelerated Quality-Diversity through Massive Parallelismcited by 14
  10. 2022Path of Destruction: Learning an Iterative Level Generator Using a Small Datasetcited by 14
  11. 2023The Five-Dollar Model: Generating Game Maps and Sprites from Sentence Embeddingscited by 14
  12. 2022Goal-Guided Neural Cellular Automata: Learning to Control Self-Organising Systemscited by 9
23 more
  1. 2022Mutation Models: Learning to Generate Levels by Imitating Evolutioncited by 9
  2. 2023Arbitrarily Scalable Environment Generators via Neural Cellular Automatacited by 9
  3. 2023Surrogate Assisted Generation of Human-Robot Interaction Scenarioscited by 8
  4. 2022Mario Plays on a Manifold: Generating Functional Content in Latent Space through Differential Geometrycited by 5
  5. 2024Measuring Diversity of Game Scenarioscited by 5
  6. 2025LatticeWorld: A Multimodal Large Language Model-Empowered Framework for Interactive Complex World Generationcited by 5
  7. 2022A Unified Substrate for Body-Brain Co-evolutioncited by 4
  8. 2022GriddlyJS: A Web IDE for Reinforcement Learningcited by 4
  9. 2023Hierarchically Composing Level Generators for the Creation of Complex Structurescited by 4
  10. 2022Augmentative Topology Agents For Open-Ended Learningcited by 3
  11. 2023Pathfinding Neural Cellular Automatacited by 3
  12. 2023Don't Bet on Luck Alone: Enhancing Behavioral Reproducibility of Quality-Diversity Solutions in Uncertain Domainscited by 3
  13. 2022Start Small: Training Controllable Game Level Generators without Training Data by Learning at Multiple Sizescited by 2
  14. 2022Efficient Learning of Locomotion Skills through the Discovery of Diverse Environmental Trajectory Generator Priorscited by 2
  15. 2023Quality Diversity in the Amorphous Fortress (QD-AF): Evolving for Complexity in 0-Player Gamescited by 1
  16. 2024Multi-Texture Synthesis through Signal Responsive Neural Cellular Automatacited by 1
  17. 2023Evolutionary Machine Learning and Gamescited by 0
  18. 2024From Metrics to Meaning: Time to Rethink Evaluation in Human-AI Collaborative Designcited by 0
  19. 2024Future Research Avenues for Artificial Intelligence in Digital Gaming: An Exploratory Reportcited by 0
  20. 2025Missing Data Imputation using Neural Cellular Automatacited by 0
  21. 2026Generating Physically Sound Designs from Text and a Set of Physical Constraintscited by 0
  22. 2026Programmable Cellular Automatacited by 0
  23. 2026Neural Cellular Automata Learn General Features in their Hidden Channelscited by 0