Academic Journal
A deep generative approach to personalized super mario level design.
| Τίτλος: | A deep generative approach to personalized super mario level design. |
|---|---|
| Συγγραφείς: | Baharvand D; Department of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran., Saeedi N; Department of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran., Gharehveran SS; Department of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran., Shirini K; Faculty of Multimedia, Tabriz Islamic Art University, Tabriz, Iran. ki.shirini@tabriziau.ac.ir. |
| Πηγή: | Scientific reports [Sci Rep] 2026 Apr 18; Vol. 16 (1). Date of Electronic Publication: 2026 Apr 18. |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: Nature Publishing Group Country of Publication: England NLM ID: 101563288 Publication Model: Electronic Cited Medium: Internet ISSN: 2045-2322 (Electronic) Linking ISSN: 20452322 NLM ISO Abbreviation: Sci Rep Subsets: MEDLINE |
| Imprint Name(s): | Original Publication: London : Nature Publishing Group, copyright 2011- |
| Ιατρικοί όροι (MeSH): | Generative Adversarial Networks* , Deep Learning* , Clustering Algorithms*, Humans ; Neural Networks, Computer ; Video Games |
| Περίληψη: | Designing game levels that appropriately match a player's skill level remains a fundamental challenge in procedural content generation, as mismatches in difficulty can lead to player boredom, frustration, and reduced engagement. While deep generative models enable automatic level synthesis at scale, the comparative effectiveness of different GAN architectures for skill-conditioned and personalized level generation remains insufficiently explored. In this work, we investigate the use of generative adversarial networks (GANs) for skill-conditioned procedural content generation by evaluating five architectures: U-Net GAN, StyleGAN, Deep Convolutional GAN (DCGAN), ResNet-GAN, and Spectral Normalization GAN (SN-GAN). Player behavior is clustered into discrete skill groups using Spectral Clustering, and the resulting labels are employed as conditioning signals for level generation. The selected architectures span different convolutional depths, normalization strategies, and regularization mechanisms, enabling a broad and systematic comparison. All models are evaluated using consistent quantitative metrics, including tile distribution entropy, diversity score, discriminator accuracy, generation speed, and pairwise Hamming distance. Quantitative analysis demonstrates that as player skill increases, generated levels exhibit fewer deaths and shorter completion times, while encouraging more complex interactions such as higher jumps and coin collection. Experimental results indicate that ResNet-based and U-Net-style GAN architectures achieve the most favorable balance between level diversity, playability, and training stability, corroborating both qualitative and quantitative assessments. The spectral clustering-based skill-conditioning approach successfully produced personalized levels aligned with each player's abilities, supporting adaptive procedural content generation across different skill groups. (© 2026. The Author(s).) |
| Competing Interests: | Declarations. Competing interests: The authors declare no competing interests. Ethical and informed consent for data used: This article does not contain any studies with human participants or animals performed by any of the authors. |
| References: | Volz, V. et al. Evolving mario levels in the latent space of a deep convolutional generative adversarial network. In Proceedings of the Genetic and Evolutionary Computation Conference (2018). Shaker, N., Togelius, J. & Nelson, M. J. Procedural Content Generation Games (2016). Shaker, N., Yannakakis, G. & Togelius, J. Towards automatic personalized content generation for platform games. In Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (2010). Lucic, M. et al. Are gans created equal? a large-scale study. Adv. Neural. Inf. Process. Syst. 31. (2018). Compton, K. & Mateas, M. Procedural level design for platform games. In Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (2006). Togelius, J. et al. Search-based procedural content generation. In European Conference on the Applications of Evolutionary Computation (Springer, 2010). Hunicke, R. The case for dynamic difficulty adjustment in games. In Proceedings of the 2005 ACM SIGCHI International Conference on Advances in Computer Entertainment Technology (2005). Csikszentmihalyi, M. & Csikzentmihaly, M. Flow: The Psychology of Optimal Experience, vol. 1990 (Harper & Row, 1990). Sweetser, P. & Wyeth, P. GameFlow: a model for evaluating player enjoyment in games. Comput. Entertain. CIE. 3 (3), 3–3 (2005). (PMID: 10.1145/1077246.1077253) Samadi Gharehveran, S., Najafpour, N. & Shirini, K. A hybrid approach for vehicle detection and tracking in low-visibility conditions based on AWBLP, YOLOv8, and GM-PHD algorithms. J. Mach. Vis. Image Process. 1–11 (2025). Ng, A., Jordan, M. & Weiss, Y. On spectral clustering: Analysis and an algorithm. Adv. Neural. Inf. Process. Syst. 14 (2001). Nemati, R., Shirini, K. & Gharehveran, S. S. FER-HA: A hybrid attention model for facial emotion recognition: R. Nemati et al. J. Supercomput. 81 (16), 1485 (2025). (PMID: 10.1007/s11227-025-07983-4) Samadi Gharehveran, S. et al. A review of game theory-based approaches for demand side management in smart energy grids. Power Control Data Process. Syst. e730417 (2025). صمدی قره & ورن س., A Review of Resilient-oriented Operation Strategies of Electrical Energy Distribution Networks in Critical Situations. پدافند غیرعامل, 2025(مقالات آماده انتشار). Gharehveran, S. S., Shirini, K. & Abdolahi, A. Energy storage devices-A comprehensive overview. Optim. Energy Storage. 47 (2025). Gharehveran, S. S. et al. Optimizing day-ahead power scheduling: a novel MIQCP approach for enhanced SCUC with renewable integration. In e-Prime-Advances in Electrical Engineering, Electronics and Energy, 101022 (2025). Saeedi, N. et al. Prediction of electrical energy consumption using principal component analysis and independent components analysis. J. Supercomput. 81 (9), 1072 (2025). (PMID: 10.1007/s11227-025-07505-2) Shirini, K., Hajivand, A. T. & Ghareveran, S. S. A novel deep learning-based method for potato leaf disease classification. 9th Adv. Eng. Days. 9, 462–464 (2024). Shirini, K. & Samadi Gharehveran, S. Balancing time and cost in resource-constrained project scheduling using meta-heuristic approach. J. Agric. Mach. 14(2). (2024). Shirini, K. & Samadi, G. S. A review on algorithms for solving the resource-constrained projects scheduling problems with considering agricultural problems. (2023). Yannakakis, G. N. & Togelius, J. Artificial Intelligence and Games, vol. 2 (Springer, 2018). Goodfellow, I. J. et al. Generative adversarial nets. Adv. Neural. Inf. Process. Syst. 27. (2014). Liu, M. Y., Breuel, T. & Kautz, J. Unsupervised image-to-image translation networks. Adv. Neural. Inf. Process. Syst. 30 (2017). Rajebi, S., Pedrammehr, S. & Shirini, K. Hybrid intelligent optimization of a circularly polarized microstrip antenna array for safe and effective hyperthermia cancer therapy. Sci. Rep. (2026). Akan, T. et al. Battle royale optimizer with ring neighborhood topology. ICCK Trans. Swarm Evol. Learn. 2 (1), 19–40 (2026). Hayati, A., Gharehveran, S. S. & Shirini, K. Electricity price forecasting with ensemble meta-models and SHAP explainers: a PCA-driven approach. Sci. Rep. (2026). Samadi Gharehveran, S. & Shirini, K. A review of resilient-oriented operation strategies of electrical energy distribution networks in critical situations. Passive Defense (2025). Taherihajivand, A., Shirini, K. & Samadi Ghareveran, S. Application of the Internet of Things in agriculture to diagnose plant diseases. Agric. Inform. Sci. Technol. 8 (2), 59–71 (2025). Togelius, J. et al. Procedural Content Generation: Goals, Challenges and Actionable Steps (Schloss Dagstuhl–Leibniz-Zentrum fuer Informatik, 2013). Fontaine, M. C. et al. Illuminating mario scenes in the latent space of a generative adversarial network. In Proceedings of the AAAI Conference on Artificial Intelligence (2021). Smith, J. B. et al. The CrossSong Puzzle: Developing a logic puzzle for musical thinking. J. New. Music Res. 46 (3), 213–228 (2017). (PMID: 10.1080/09298215.2017.1303519) Baldwin, A. et al. Towards pattern-based mixed-initiative dungeon generation. In Proceedings of the 12th International Conference on the Foundations of Digital Games (2017). Wang, D. et al. Frequency-to‐spectrum mapping GAN for semisupervised hyperspectral anomaly detection. CAAI Trans. Intell. Technol. 8 (4), 1258–1273 (2023). (PMID: 10.1049/cit2.12154) Tashi, D. et al. Tibetan Data Augmentation via GAN-Based Handwritten Text Generation (CAAI Transactions on Intelligence Technology, 2025). Petzka, H., Fischer, A. & Lukovnicov, D. On the regularization of wasserstein gans. arXiv preprint arXiv:1709.08894 (2017). He, K. et al. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2016). Karras, T., Laine, S. & Aila, T. A style-based generator architecture for generative adversarial networks. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2019). Liu, M. Y. & Tuzel, O. Coupled generative adversarial networks. Adv. Neural. Inf. Process. Syst. 29 (2016). Heusel, M. et al. Gans trained by a two time-scale update rule converge to a local nash equilibrium. Adv. Neural. Inf. Process. Syst. 30 (2017). Sattari, M. T., Shirini, K. & Javidan, S. Evaluating the efficiency of dimensionality reduction methods in improving the accuracy of water quality index modeling in Qizil-Uzen River using machine learning algorithms. Water Soil. Manag. Model. 4 (2), 89–104 (2024). Shirini, K., Aghdasi, H. S. & Saeedvand, S. Modified imperialist competitive algorithm for aircraft landing scheduling problem: K. Shirini. J. Supercomput. 80 (10), 13782–13812 (2024). (PMID: 10.1007/s11227-024-05999-w) Zaki Dizaji, H. et al. Modelling variables affecting the yield of sugarcane fields using deep recurrent neural network. Iran. J. Biosyst. Eng. 55 (2), 93–108 (2024). Summerville, A. J. et al. The vglc: The video game level corpus. arXiv preprint arXiv:1606.07487, (2016). Radford, A., Metz, L. & Chintala, S. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:1511.06434, (2015). Pedregosa, F. et al. Scikit-learn: Machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011). MacQueen, J. Multivariate observations. In Proceedings ofthe 5th Berkeley Symposium on Mathematical Statisticsand Probability (1967). Ester, M. et al. A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise. In kdd (1996). Gholinavaz, S., Saeedi, N. & Gharehveran, S. S. Robustness analysis of YOLO and faster R-CNN for object detection in realistic weather scenarios with noise augmentation. Sci. Rep. 15 (1), 44888 (2025). (PMID: 10.1038/s41598-025-28737-54146174212749240) Kaufman, L. & Rousseeuw, P. J. Finding Groups in Data: An Introduction to Cluster Analysis (Wiley, 2009). McLachlan, G. J. & Peel, D. Finite Mixture Models (Wiley, 2000). Von Luxburg, U. A tutorial on spectral clustering. Stat. Comput. 17 (4), 395–416 (2007). (PMID: 10.1007/s11222-007-9033-z) Ronneberger, O., Fischer, P. & Brox, T. U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention (Springer, 2015). Shannon, C. E. A mathematical theory of communication. Bell Syst. Tech. J. 27 (3), 379–423 (1948). (PMID: 10.1002/j.1538-7305.1948.tb01338.x) Dempster, A. P., Laird, N. M. & Rubin, D. B. Maximum likelihood from incomplete data via the EM algorithm. J. R. Stat. Soc. Ser. B (Methodol.). 39 (1), 1–22 (1977). (PMID: 10.1111/j.2517-6161.1977.tb01600.x) Dempster, A. Maximum likelihood estimation from incomplete data via the EM algorithm. J. R. Stat. Soc. 39, 1–38 (1977). (PMID: 10.1111/j.2517-6161.1977.tb01600.x) Caliński, T. & Harabasz, J. A dendrite method for cluster analysis. Commun. Stat.-Theory Methods. 3 (1), 1–27 (1974). Rousseeuw, P. J. Silhouettes: A graphical aid to the interpretation and validation of cluster analysis. J. Comput. Appl. Math. 20, 53–65 (1987). (PMID: 10.1016/0377-0427(87)90125-7) Davies, D. L. & Bouldin, D. W. A cluster separation measure. IEEE Trans. Pattern Anal. Mach. Intell. 2009(2), 224–227 (2009). Arbelaitz, O. et al. An extensive comparative study of cluster validity indices. Pattern Recogn. 46 (1), 243–256 (2013). (PMID: 10.1016/j.patcog.2012.07.021) Jain, A. K. Data clustering: 50 years beyond K-means. Pattern Recognit. Lett. 31 (8), 651–666 (2010). (PMID: 10.1016/j.patrec.2009.09.011) |
| Contributed Indexing: | Keywords: Deep generative models; Generative adversarial networks (GANs); Personalized game design; Procedural content generation (PCG); Skill-conditioned level generation |
| Entry Date(s): | Date Created: 20260418 Date Completed: 20260612 Latest Revision: 20260726 |
| Update Code: | 20260726 |
| PubMed Central ID: | PMC13254052 |
| DOI: | 10.1038/s41598-026-46199-1 |
| PMID: | 42000807 |
| Βάση Δεδομένων: | MEDLINE |
| ISSN: | 2045-2322 |
|---|---|
| DOI: | 10.1038/s41598-026-46199-1 |