2000 character limit reached
Procedural content generation of puzzle games using conditional generative adversarial networks (2306.15696v1)
Published 26 Jun 2023 in cs.AI and cs.LG
Abstract: In this article, we present an experimental approach to using parameterized Generative Adversarial Networks (GANs) to produce levels for the puzzle game Lily's Garden. We extract two condition vectors from the real levels in an effort to control the details of the GAN's outputs. While the GANs perform well in approximating the first condition (map shape), they struggle to approximate the second condition (piece distribution). We hypothesize that this might be improved by trying out alternative architectures for both the Generator and Discriminator of the GANs.
- Wasserstein GAN. (2017). arXiv:1701.07875 http://arxiv.org/abs/1701.07875
- DOOM Level Generation Using Generative Adversarial Networks. 2018 IEEE Games, Entertainment, Media Conference, GEM 2018 (2018), 316–323. https://doi.org/10.1109/GEM.2018.8516539 arXiv:1804.09154
- Ian Goodfellow. 2016. NIPS 2016 Tutorial: Generative Adversarial Networks. (dec 2016). arXiv:1701.00160 http://arxiv.org/abs/1701.00160
- Generative adversarial nets. Advances in Neural Information Processing Systems 3, January (2014), 2672–2680. https://doi.org/10.3156/jsoft.29.5_177_2 arXiv:1406.2661
- Improved training of wasserstein GANs. Advances in Neural Information Processing Systems 2017-December (2017), 5768–5778. arXiv:1704.00028
- Pei.J Han.J, Kamber.M. 2012. Data Mining: Concepts and Techniques. Elsevier, Morgan Kauffman. 1–703 pages.
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium. Nips (2017). arXiv:1706.08500 http://arxiv.org/abs/1706.08500
- Sentient Sketchbook : Computer-Assisted Game Level Authoring. ACM, Chania. https://www.um.edu.mt/library/oar/handle/123456789/29607
- Mehdi Mirza and Simon Osindero. 2014. Conditional Generative Adversarial Nets. (2014), 1–7. arXiv:1411.1784 http://arxiv.org/abs/1411.1784
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks. (nov 2015). arXiv:1511.06434 http://arxiv.org/abs/1511.06434
- Improved techniques for training GANs. Advances in Neural Information Processing Systems (2016), 2234–2242. arXiv:1606.03498
- Controllable Level Blending between Games using Variational Autoencoders. (2017). arXiv:arXiv:2002.11869v1
- Procedural Content Generation in Games. Springer Nature. 1–234 pages.
- Noor Shaker and GN Yannakakis. 2009. Towards automatic personalized content generation for platform games. Proc. 6th Artif. Intell. Interactive Digit. Hudlicka 2008 (2009). http://www.aaai.org/ocs/index.php/AIIDE/AIIDE10/paper/viewPDFInterstitial/2135/2546
- Sam Snodgrass and Santiago Onta. 2015. Controllable Procedural Content Generation via Constrained Multi-Dimensional Markov Chain Sampling. MdMC (2015), 780–786.
- Adam Summerville. 2018. Expanding Expressive Range : Evaluation Methodologies for Procedural Content Generation. AIIDE (2018), 116–122.
- Procedural Content Generation via Machine Learning (PCGML). (feb 2017). arXiv:1702.00539 http://arxiv.org/abs/1702.00539
- Towards automatic personalised content creation for racing games. In Proceedings of the 2007 IEEE Symposium on Computational Intelligence and Games, CIG 2007. 252–259. https://doi.org/10.1109/CIG.2007.368106
- Bootstrapping Conditional GANs for Video Game Level Generation. (2019). arXiv:1910.01603 http://arxiv.org/abs/1910.01603
- Evolving Mario Levels in the Latent Space of a Deep Convolutional Generative Adversarial Network. (may 2018). arXiv:1805.00728 http://arxiv.org/abs/1805.00728
Collections
Sign up for free to add this paper to one or more collections.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
