mlsanityData to 7 Oct 2026

Illuminating the Space of Beatable Lode Runner Levels Produced By Various Generative Adversarial Networks

arXiv 2101.07868 ↗First posted 19 Jan 2021cs.LG cs.AI cs.NEcited by 14

Not on the map. Why?

Abstract (arXiv)

Generative Adversarial Networks (GANs) are capable of generating convincing imitations of elements from a training set, but the distribution of elements in the training set affects to difficulty of properly training the GAN and the quality of the outputs it produces. This paper looks at six different GANs trained on different subsets of data from the game Lode Runner. The quality diversity algorithm MAP-Elites was used to explore the set of quality levels that could be produced by each GAN, where quality was defined as being beatable and having the longest solution path possible. Interestingly, a GAN trained on only 20 levels generated the largest set of diverse beatable levels while a GAN trained on 150 levels generated the smallest set of diverse beatable levels, thus challenging the notion that more is always better when training GANs.

Key facts. Cited by 14 papers in the map (data through 7 Oct 2026). Builds on 11 of its 23 references that are in the map. First posted on arXiv 19 Jan 2021.

Builds on

11 of 23 references.

References · 11

  1. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
  2. 2015Illuminating search spaces by mapping elitescited by 406
  3. 2018Evolving Mario Levels in the Latent Space of a Deep Convolutional Generative Adversarial Networkcited by 76
  4. 2020Illuminating Mario Scenes in the Latent Space of a Generative Adversarial Networkcited by 51
  5. 2016The VGLC: The Video Game Level Corpuscited by 41
  6. 2019Bootstrapping Conditional GANs for Video Game Level Generationcited by 35
  7. 2020Interactive Evolution and Exploration Within Latent Level-Design Space of Generative Adversarial Networkscited by 22
  8. 2020TOAD-GAN: Coherent Style Level Generation from a Single Examplecited by 22
  9. 2020CPPN2GAN: Combining Compositional Pattern Producing Networks and GANs for Large-scale Pattern Generationcited by 19
  10. 2020Generative Adversarial Network Rooms in Generative Graph Grammar Dungeons for The Legend of Zeldacited by 8
  11. 2017DeepMasterPrints: Generating MasterPrints for Dictionary Attacks via Latent Variable Evolutioncited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 3; 2022: 4; 2023: 4; 2024: 1; 2025: 0; 2026: 2.

14 papers.

Sort:

  1. 2021Differentiable Quality Diversitycited by 71
  2. 2022Deep Surrogate Assisted Generation of Environmentscited by 40
  3. 2022Covariance Matrix Adaptation MAP-Annealingcited by 24
  4. 2023Proximal Policy Gradient Arborescence for Quality Diversity Reinforcement Learningcited by 20
  5. 2021Dungeon and Platformer Level Blending and Generation using Conditional VAEscited by 10
  6. 2023Surrogate Assisted Generation of Human-Robot Interaction Scenarioscited by 8
  7. 2022Online Damage Recovery for Physical Robots with Hierarchical Quality-Diversitycited by 7
  8. 2023Lode Encoder: AI-constrained co-creativitycited by 6
  9. 2022Hierarchical Quality-Diversity for Online Damage Recoverycited by 5
  10. 2024Measuring Diversity of Game Scenarioscited by 5
  11. 2021Using Multiple Generative Adversarial Networks to Build Better-Connected Levels for Mega Mancited by 3
  12. 2023Lode Enhancer: Level Co-creation Through Scalingcited by 1
2 more
  1. 2026Discount Model Search for Quality Diversity Optimization in High-Dimensional Measure Spacescited by 1
  2. 2026Learning Local Constraints for Reinforcement-Learned Content Generatorscited by 0