Graph Representation Learning for Game Provenance
Resumo
Introdução: Grafos de Proveniência de Jogos modelam sessões de jogos ao capturar elementos de jogos, seus estados e interações, refletindo a heterogeneidade, dinamicidade e espacialidade de seu domínio. Redes neurais de grafos (GNNs), amplamente usadas no aprendizado de máquina, aprendem representações vetoriais a partir de estruturas de grafos. Embora trabalhos anteriores tenham tentado combinar GNNs com a modelagem de grafos específicos a domínios, estes normalmente negligenciam a inerente heterogeneidade do domínio dos jogos. Objetivo: A tese tem como objetivo identificar os principais desafio na integração de Grafos de Proveniência de Jogos ao Aprendizado de Representação de grafos, introduzir uma solução que lida explicitamente a natureza heterogênea dos Grafos de Proveniência de Jogos, e avaliar sua efetividade em tarefas de Game Analytics. Metodologia ou Passos: Para alcançar esses objetivos, a tese revisita os conceitos principais de Grafos de Proveniência e Aprendizado de Representação baseado em GNN, revisa abordagens existentes na literatura e aplicações em jogos digitais, introduz o framework PinGLR, e apresenta um novo dataset de grafos de proveniência. Resultados: As contribuições incluem: (i) o framework PinGLR, que aborda a heterogeneidade e oferece suporte a pipelines completos de aprendizado de máquina, (ii) um método inovador para lidar com a heterogeneidade dos nós por meio da interseção recorrente de conjuntos de atributos, e (iii) evidências empíricas demonstrando que o PinGLR alcança desempenho comparável ou superior ao de modelos tradicionais em tarefas de Game Analytics.
Palavras-chave:
Aprendizado de Representações, Grafo de Proveniência de Jogos, Embeddings de Grafos, Redes Neurais de Grafos, Game Analytics
Referências
Alencar, V. B. (2020). Prov-dominoes: an exploratory analysis approach for provenance data. Master’s thesis, Universidade Federal Fluminense, Niterói.
Angles, R. e Gutierrez, C. (2008). Survey of graph database models. ACM Computing Surveys (CSUR), 40(1):1–39.
Bengio, Y., Courville, A., e Vincent, P. (2013). Representation learning: A review and new perspectives. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(8):1798–1828.
Cen, Y., Zou, X., Zhang, J., Yang, H., Zhou, J., e Tang, J. (2019). Representation learning for attributed multiplex heterogeneous network. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pages 1358–1368.
Chen, F., Wang, Y.-C., Wang, B., e Kuo, C.-C. J. (2020). Graph representation learning: a survey. APSIPA Transactions on Signal and Information Processing, 9:e15.
Drachen, A., Seif El-Nasr, M., e Canossa, A. (2013). Game analytics–the basics. Game analytics: Maximizing the value of player data, pages 13–40.
Figueira, F. M., Nascimento, L., da Silva Junior, J., Kohwalter, T., Murta, L., e Clua, E. (2018). Bing: A framework for dynamic game balancing using provenance. In Proc. of the 17th Brazilian Symposium on Computer Games and Digital Entertainment (SBGames), pages 57–5709. IEEE.
Gilmer, J., Schoenholz, S. S., Riley, P. F., Vinyals, O., e Dahl, G. E. (2017). Neural message passing for quantum chemistry. In Proceedings of the 34th International Conference on Machine Learning - Volume 70, ICML’17, page 1263–1272. JMLR.org.
Hamilton, W. L., Ying, R., e Leskovec, J. (2017). Representation learning on graphs: Methods and applications. IEEE Data Eng. Bull., 40(3):52–74.
Kazemi, S. M., Goel, R., Jain, K., Kobyzev, I., Sethi, A., Forsyth, P., e Poupart, P. (2020). Representation learning for dynamic graphs: A survey. Journal of Machine Learning Research, 21(70):1–73.
Ke, C. H., Deng, H., Xu, C., Li, J., Gu, X., Yadamsuren, B., Klabjan, D., Sifa, R., Drachen, A., e Demediuk, S. (2022). Dota 2 match prediction through deep learning team fight models. In 2022 IEEE Conference on Games (CoG), pages 96–103.
Kohwalter, T., Clua, E., e Murta, L. (2012). Provenance in games. XI SBGames, pages 162–171.
Kohwalter, T., Oliveira, T., Freire, J., Clua, E., e Murta, L. (2016). Prov viewer: A graph-based visualization tool for interactive exploration of provenance data. In International Provenance and Annotation Workshop, pages 71–82. Springer.
Kohwalter, T. C., de Azeredo Figueira, F. M., de Lima Serdeiro, E. A., da Silva Junior, J. R., Murta, L. G. P., e Clua, E. W. G. (2018). Understanding game sessions through provenance. Entertainment Computing, 27:110–127.
Kohwalter, T. C., Murta, L. G. P., e Clua, E. W. G. (2017). Capturing game telemetry with provenance. In 2017 16th Brazilian Symposium on Computer Games and Digital Entertainment (SBGames), pages 66–75. IEEE.
Le-Khac, P. H., Healy, G., e Smeaton, A. F. (2020). Contrastive representation learning: A framework and review. IEEE Access, 8:193907–193934.
Melo, S. A. (2019). Detecting long-range cause-effect relationships in game provenance graphs with graph-based representation learning. Master’s thesis, Universidade Federal Fluminense, Niterói.
Melo, S. A., Kohwalter, T. C., Clua, E., Paes, A., e Murta, L. (2020). Player behavior profiling through provenance graphs and representation learning. pages 62:1–62:11.
Melo, S. A., Paes, A., Clua, E. W. G., Kohwalter, T. C., e Murta, L. G. P. (2019). Detecting long-range cause-effect relationships in game provenance graphs with graph-based representation learning. Entertainment Computing, 32:100318.
Moreau, L., Clifford, B., Freire, J., Futrelle, J., Gil, Y., Groth, P., Kwasnikowska, N., Miles, S., Missier, P., Myers, J., et al. (2011). The open provenance model core specification (v1. 1). Future generation computer systems, 27(6):743–756.
Myers, S. A., Sharma, A., Gupta, P., e Lin, J. (2014). Information network or social network? the structure of the twitter follow graph. In Proceedings of the 23rd International Conference on World Wide Web, pages 493–498.
Skarding, J., Gabrys, B., e Musial, K. (2021). Foundations and modeling of dynamic networks using dynamic graph neural networks: A survey. IEEE Access, 9:79143–79168.
Wang, X., Bo, D., Shi, C., Fan, S., Ye, Y., e Yu, P. S. (2020). A survey on heterogeneous graph embedding: Methods, techniques, applications and sources. arXiv preprint arXiv:2011.14867.
Zhang, C., Song, D., Huang, C., Swami, A., e Chawla, N. V. (2019). Heterogeneous graph neural network. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pages 793–803.
Angles, R. e Gutierrez, C. (2008). Survey of graph database models. ACM Computing Surveys (CSUR), 40(1):1–39.
Bengio, Y., Courville, A., e Vincent, P. (2013). Representation learning: A review and new perspectives. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(8):1798–1828.
Cen, Y., Zou, X., Zhang, J., Yang, H., Zhou, J., e Tang, J. (2019). Representation learning for attributed multiplex heterogeneous network. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pages 1358–1368.
Chen, F., Wang, Y.-C., Wang, B., e Kuo, C.-C. J. (2020). Graph representation learning: a survey. APSIPA Transactions on Signal and Information Processing, 9:e15.
Drachen, A., Seif El-Nasr, M., e Canossa, A. (2013). Game analytics–the basics. Game analytics: Maximizing the value of player data, pages 13–40.
Figueira, F. M., Nascimento, L., da Silva Junior, J., Kohwalter, T., Murta, L., e Clua, E. (2018). Bing: A framework for dynamic game balancing using provenance. In Proc. of the 17th Brazilian Symposium on Computer Games and Digital Entertainment (SBGames), pages 57–5709. IEEE.
Gilmer, J., Schoenholz, S. S., Riley, P. F., Vinyals, O., e Dahl, G. E. (2017). Neural message passing for quantum chemistry. In Proceedings of the 34th International Conference on Machine Learning - Volume 70, ICML’17, page 1263–1272. JMLR.org.
Hamilton, W. L., Ying, R., e Leskovec, J. (2017). Representation learning on graphs: Methods and applications. IEEE Data Eng. Bull., 40(3):52–74.
Kazemi, S. M., Goel, R., Jain, K., Kobyzev, I., Sethi, A., Forsyth, P., e Poupart, P. (2020). Representation learning for dynamic graphs: A survey. Journal of Machine Learning Research, 21(70):1–73.
Ke, C. H., Deng, H., Xu, C., Li, J., Gu, X., Yadamsuren, B., Klabjan, D., Sifa, R., Drachen, A., e Demediuk, S. (2022). Dota 2 match prediction through deep learning team fight models. In 2022 IEEE Conference on Games (CoG), pages 96–103.
Kohwalter, T., Clua, E., e Murta, L. (2012). Provenance in games. XI SBGames, pages 162–171.
Kohwalter, T., Oliveira, T., Freire, J., Clua, E., e Murta, L. (2016). Prov viewer: A graph-based visualization tool for interactive exploration of provenance data. In International Provenance and Annotation Workshop, pages 71–82. Springer.
Kohwalter, T. C., de Azeredo Figueira, F. M., de Lima Serdeiro, E. A., da Silva Junior, J. R., Murta, L. G. P., e Clua, E. W. G. (2018). Understanding game sessions through provenance. Entertainment Computing, 27:110–127.
Kohwalter, T. C., Murta, L. G. P., e Clua, E. W. G. (2017). Capturing game telemetry with provenance. In 2017 16th Brazilian Symposium on Computer Games and Digital Entertainment (SBGames), pages 66–75. IEEE.
Le-Khac, P. H., Healy, G., e Smeaton, A. F. (2020). Contrastive representation learning: A framework and review. IEEE Access, 8:193907–193934.
Melo, S. A. (2019). Detecting long-range cause-effect relationships in game provenance graphs with graph-based representation learning. Master’s thesis, Universidade Federal Fluminense, Niterói.
Melo, S. A., Kohwalter, T. C., Clua, E., Paes, A., e Murta, L. (2020). Player behavior profiling through provenance graphs and representation learning. pages 62:1–62:11.
Melo, S. A., Paes, A., Clua, E. W. G., Kohwalter, T. C., e Murta, L. G. P. (2019). Detecting long-range cause-effect relationships in game provenance graphs with graph-based representation learning. Entertainment Computing, 32:100318.
Moreau, L., Clifford, B., Freire, J., Futrelle, J., Gil, Y., Groth, P., Kwasnikowska, N., Miles, S., Missier, P., Myers, J., et al. (2011). The open provenance model core specification (v1. 1). Future generation computer systems, 27(6):743–756.
Myers, S. A., Sharma, A., Gupta, P., e Lin, J. (2014). Information network or social network? the structure of the twitter follow graph. In Proceedings of the 23rd International Conference on World Wide Web, pages 493–498.
Skarding, J., Gabrys, B., e Musial, K. (2021). Foundations and modeling of dynamic networks using dynamic graph neural networks: A survey. IEEE Access, 9:79143–79168.
Wang, X., Bo, D., Shi, C., Fan, S., Ye, Y., e Yu, P. S. (2020). A survey on heterogeneous graph embedding: Methods, techniques, applications and sources. arXiv preprint arXiv:2011.14867.
Zhang, C., Song, D., Huang, C., Swami, A., e Chawla, N. V. (2019). Heterogeneous graph neural network. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pages 793–803.
Publicado
30/09/2025
Como Citar
MELO, Sidney; PAES, Aline; CLUA, Esteban.
Graph Representation Learning for Game Provenance. In: CONCURSO DE TESES E DISSERTAÇÕES – DOUTORADO - SIMPÓSIO BRASILEIRO DE JOGOS E ENTRETENIMENTO DIGITAL (SBGAMES), 14. , 2025, Salvador/BA.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 364-370.
DOI: https://doi.org/10.5753/sbgames_estendido.2025.11584.
