Bootstrapping Conditional GANs for Video Game Level Generation

R. Rodriguez Torrado, A. Khalifa, Michael Cerny Green, N. Justesen, S. Risi, J. Togelius

Research output: Conference Article in Proceeding or Book/Report chapterArticle in proceedingsResearchpeer-review


Generative Adversarial Networks (GANs) have shown impressive results for image generation. However, GANs face challenges in generating contents with certain types of constraints, such as game levels. Specifically, it is difficult to generate levels that have aesthetic appeal and are playable at the same time. Additionally, because training data usually is limited, it is challenging to generate unique levels with current GANs. In this paper, we propose a new GAN architecture named Conditional Embedding Self-Attention Generative Adversarial Net-work (CESAGAN) and a new bootstrapping training procedure. The CESAGAN is a modification of the self-attention GAN that incorporates an embedding feature vector input to condition the training of the discriminator and generator. This allows the network to model non-local dependency between game objects, and to count objects. Additionally, to reduce the number of levels necessary to train the GAN, we propose a bootstrapping mechanism in which playable generated levels are added to the training set. The results demonstrate that the new approach does not only generate a larger number of levels that are playable but also generates fewer duplicate levels compared to a standard GAN.
Original languageUndefined/Unknown
Title of host publication2020 IEEE Conference on Games (CoG)
Number of pages8
Publication date2020
ISBN (Electronic)978-1-7281-4533-4
Publication statusPublished - 2020


  • Generative Adversarial Networks
  • Image Generation
  • Game Level Generation
  • Conditional Embedding
  • Self-Attention
  • Non-Local Dependency
  • Bootstrapping Training
  • Playable Content Generation
  • Aesthetic Appeal
  • Training Data Augmentation

Cite this