2D Map Generation for Games Using Generative Adversarial Networks
Resumo
Introduction: Video games have evolved from simple pastimes into a culturally significant art form, driven by technological advances that have enabled more realistic graphics and immersive gameplay. Objective: This study explores the use of Generative Adversarial Networks (GANs) for generating maps in 2D games, focusing on Super Mario Bros. Methodology: The methodology includes the use of a Wasserstein GAN (WGAN), combined with a fragment selection algorithm and gameplay evaluation performed by an A* agent. Results: The results indicate a significant improvement in map quality, with a 25% increase in the rate of playable fragments and a 15% reduction in the agent’s average evaluation time, ensuring greater adherence to the characteristics of the original game.
Palavras-chave:
Map Generation, Geração de mapa, GANs, Super Mario Bros, PCG
Referências
Aloupis, G., Demaine, E. D., Guo, A., e Viglietta, G. (2015). Classic nintendo games are (computationally) hard. Theoretical Computer Science, 586:135–160. Fun with Algorithms.
Andrade, E. S. (2023). Games e a indústria cultural: O impacto sociocultural dos jogos eletrônicos. Repositório Institucional - Universidade Federal de Uberlândia: Home.
Arjovsky, M., Chintala, S., e Bottou, L. (2017). Wasserstein generative adversarial networks. In Precup, D. e Teh, Y. W., editors, Proceedings of the 34th International Conference on Machine Learning, volume 70 of Proceedings of Machine Learning Research, pages 214–223. PMLR.
Awiszus, M., Schubert, F., e Rosenhahn, B. (2020). Toad-gan: Coherent style level generation from a single example. ArXiv.
Cheigh, J. (2023). Generating images using vaes, gans, and diffusion models. [link]. [Accessed 09-03-2024].
da Manhã, D. (2023). 4 importantes transformações tecnológicas nos jogos eletrônicos. [link]. [Accessed 19-02-2024].
David J. Malan, D. L. (2021). Csci e-23a - cs50. [link]. [Accessed 09-03-2024].
deWinter, J. (2015). Shigeru Miyamoto: Super Mario Bros., Donkey Kong, The Legend of Zelda. Influential Video Game Designers. Bloomsbury Publishing.
e Silva, D. F., Torchelsen, R., e Aguiar, M. (2023). Dungeon level generation using generative adversarial network: an experimental study for top-down view games. In Anais do L Seminário Integrado de Software e Hardware, pages 95–106, Porto Alegre, RS, Brasil. SBC.
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., e Bengio, Y. (2020). Generative adversarial networks. Commun. ACM, 63(11):139–144.
Gulrajani, I., Ahmed, F., Arjovsky, M., Dumoulin, V., e Courville, A. (2017). Improved training of wasserstein gans. In Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS’17, page 5769–5779, Red Hook, NY, USA. Curran Associates Inc.
Khalifa, A. (2021). Mario-ai-framework. [link]. [Accessed 27-07-2024].
Kingma, D. P. e Welling, M. (2019). An introduction to variational autoencoders. Foundations and Trends® in Machine Learning, 12(4):307–392.
Le Roy Matos, V. (2025). mario-maps-generation-gans-selection: 2d map generation for games using generative adversarial networks. [link].
Machado, R. P. T. (2022). Jogos eletrônicos e e-sports: Um fenômeno cultural. [link]. [Accessed 19-02-2024].
Reis, M. d. L. e Andrade, K. d. O. (2023). Áreas de pesquisa e técnicas de inteligência artificial em jogos digitais. Revista Tecnológica Fatec Americana.
Schrum, J., Gutierrez, J., Volz, V., Liu, J., Lucas, S., e Risi, S. (2020). Interactive evolution and exploration within latent level-design space of generative adversarial networks. In Proceedings of the 2020 Genetic and Evolutionary Computation Conference, GECCO ’20, page 148–156, New York, NY, USA. Association for Computing Machinery.
Shaham, T., Dekel, T., e Michaeli, T. (2019). Singan: Learning a generative model from a single natural image. In 2019 IEEE/CVF International Conference on Computer Vision (ICCV), pages 4569–4579, Los Alamitos, CA, USA. IEEE Computer Society.
Shaker, N., Togelius, J., Yannakakis, G. N., Weber, B., Shimizu, T., Hashiyama, T., Sorenson, N., Pasquier, P., Mawhorter, P., Takahashi, G., Smith, G., e Baumgarten, R. (2011). The 2010 mario ai championship: Level generation track. IEEE Transactions on Computational Intelligence and AI in Games, 3(4):332–347.
Siqueira, M. S. (2023). Esports e o ambiente acadêmico: uma perspectiva sobre o cenário esports. Revista UFF.
Summerville, A. J., Snodgrass, S., Mateas, M., e Villar, S. O. (2016). The VGLC: The video game level corpus. CoRR, abs/1606.07487.
Togelius, J., Karakovskiy, S., e Baumgarten, R. (2010). The 2009 mario ai competition. In IEEE Congress on Evolutionary Computation, pages 1–8.
Volz, V., Schrum, J., Liu, J., Lucas, S. M., Smith, A., e Risi, S. (2018). Evolving mario levels in the latent space of a deep convolutional generative adversarial network. In Proceedings of the Genetic and Evolutionary Computation Conference, GECCO ’18, page 221–228, New York, NY, USA. Association for Computing Machinery.
Andrade, E. S. (2023). Games e a indústria cultural: O impacto sociocultural dos jogos eletrônicos. Repositório Institucional - Universidade Federal de Uberlândia: Home.
Arjovsky, M., Chintala, S., e Bottou, L. (2017). Wasserstein generative adversarial networks. In Precup, D. e Teh, Y. W., editors, Proceedings of the 34th International Conference on Machine Learning, volume 70 of Proceedings of Machine Learning Research, pages 214–223. PMLR.
Awiszus, M., Schubert, F., e Rosenhahn, B. (2020). Toad-gan: Coherent style level generation from a single example. ArXiv.
Cheigh, J. (2023). Generating images using vaes, gans, and diffusion models. [link]. [Accessed 09-03-2024].
da Manhã, D. (2023). 4 importantes transformações tecnológicas nos jogos eletrônicos. [link]. [Accessed 19-02-2024].
David J. Malan, D. L. (2021). Csci e-23a - cs50. [link]. [Accessed 09-03-2024].
deWinter, J. (2015). Shigeru Miyamoto: Super Mario Bros., Donkey Kong, The Legend of Zelda. Influential Video Game Designers. Bloomsbury Publishing.
e Silva, D. F., Torchelsen, R., e Aguiar, M. (2023). Dungeon level generation using generative adversarial network: an experimental study for top-down view games. In Anais do L Seminário Integrado de Software e Hardware, pages 95–106, Porto Alegre, RS, Brasil. SBC.
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., e Bengio, Y. (2020). Generative adversarial networks. Commun. ACM, 63(11):139–144.
Gulrajani, I., Ahmed, F., Arjovsky, M., Dumoulin, V., e Courville, A. (2017). Improved training of wasserstein gans. In Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS’17, page 5769–5779, Red Hook, NY, USA. Curran Associates Inc.
Khalifa, A. (2021). Mario-ai-framework. [link]. [Accessed 27-07-2024].
Kingma, D. P. e Welling, M. (2019). An introduction to variational autoencoders. Foundations and Trends® in Machine Learning, 12(4):307–392.
Le Roy Matos, V. (2025). mario-maps-generation-gans-selection: 2d map generation for games using generative adversarial networks. [link].
Machado, R. P. T. (2022). Jogos eletrônicos e e-sports: Um fenômeno cultural. [link]. [Accessed 19-02-2024].
Reis, M. d. L. e Andrade, K. d. O. (2023). Áreas de pesquisa e técnicas de inteligência artificial em jogos digitais. Revista Tecnológica Fatec Americana.
Schrum, J., Gutierrez, J., Volz, V., Liu, J., Lucas, S., e Risi, S. (2020). Interactive evolution and exploration within latent level-design space of generative adversarial networks. In Proceedings of the 2020 Genetic and Evolutionary Computation Conference, GECCO ’20, page 148–156, New York, NY, USA. Association for Computing Machinery.
Shaham, T., Dekel, T., e Michaeli, T. (2019). Singan: Learning a generative model from a single natural image. In 2019 IEEE/CVF International Conference on Computer Vision (ICCV), pages 4569–4579, Los Alamitos, CA, USA. IEEE Computer Society.
Shaker, N., Togelius, J., Yannakakis, G. N., Weber, B., Shimizu, T., Hashiyama, T., Sorenson, N., Pasquier, P., Mawhorter, P., Takahashi, G., Smith, G., e Baumgarten, R. (2011). The 2010 mario ai championship: Level generation track. IEEE Transactions on Computational Intelligence and AI in Games, 3(4):332–347.
Siqueira, M. S. (2023). Esports e o ambiente acadêmico: uma perspectiva sobre o cenário esports. Revista UFF.
Summerville, A. J., Snodgrass, S., Mateas, M., e Villar, S. O. (2016). The VGLC: The video game level corpus. CoRR, abs/1606.07487.
Togelius, J., Karakovskiy, S., e Baumgarten, R. (2010). The 2009 mario ai competition. In IEEE Congress on Evolutionary Computation, pages 1–8.
Volz, V., Schrum, J., Liu, J., Lucas, S. M., Smith, A., e Risi, S. (2018). Evolving mario levels in the latent space of a deep convolutional generative adversarial network. In Proceedings of the Genetic and Evolutionary Computation Conference, GECCO ’18, page 221–228, New York, NY, USA. Association for Computing Machinery.
Publicado
30/09/2025
Como Citar
MATOS, Victor Le Roy; GOMES, Rogério M.; VILA NOVA, João Gabriel Gama.
2D Map Generation for Games Using Generative Adversarial Networks. In: TRILHA DE COMPUTAÇÃO – ARTIGOS CURTOS - SIMPÓSIO BRASILEIRO DE JOGOS E ENTRETENIMENTO DIGITAL (SBGAMES), 14. , 2025, Salvador/BA.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 31-37.
DOI: https://doi.org/10.5753/sbgames_estendido.2025.8849.
