Large Language Models and Dynamic Difficulty Adjustment: An Integration Perspective
Resumo
Dynamic Difficulty Adjustment (DDA) aims to enhance player retention by adjusting the difficulty level of a game according to the player's skills. However, rule-based DDA systems often struggle with scalability, adaptability, and the cognitive burden of defining exhaustive rules. In this paper, we propose the integration of Large Language Models (LLMs) into DDA mechanisms to overcome these limitations, based on prompt creation techniques. As a proof of concept, we apply this approach to the DDA-MAPEKit framework, replacing its static rule-based logic with LLM actions using ChatGPT, and an experiment was conducted in a space shooter game. The results showed promising outcomes, with pertinent adjustments to the game variables, no hallucinations, and values coupled to the current context. Overall, our findings suggest that LLM-based DDA mechanisms hold significant potential for improving adaptivity in digital games.
Palavras-chave:
Dynamic Difficulty Adjustment, Large Language Models
Referências
Bicalho, L. F., Baffa, A., e Feijó, B. (2023). A dynamic difficulty adjustment algorithm with generic player behavior classification unity plugin in single player games. In Proceedings of the 22nd Brazilian Symposium on Games and Digital Entertainment, SBGames 2023, pages 1–10, Rio Grande. ACM.
Figueira, F., Nascimento, L., Junior, J., Kohwalter, T., Murta, L., e Clua, E. (2018). BinG: A framework for dynamic game balancing using provenance. In 2018 17th SBGames, pages 57 – 66, Brazil. IEEE.
Gallotta, R., Todd, G., Zammit, M., Earle, S., Liapis, A., Togelius, J., e Yannakakis, G. N. (2024). Large language models and games: A survey and roadmap.
Hendrix, M., Bellamy-Wood, T., McKay, S., Bloom, V., e Dunwell, I. (2019). Implementing adaptive game difficulty balancing in serious games. IEEE Transactions on Games, 11(4):320–327.
Hu, S., Huang, T., Ilhan, F., Tekin, S., Liu, G., Kompella, R., e Liu, L. (2024). A survey on large language model-based game agents.
Mi, Q. e Gao, T. (2022). Improved belgian AI algorithm for dynamic management in action role-playing games. Applied Sciences, 12(22):11860.
Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., e Sutskever, I. (2019). Language models are unsupervised multitask learners.
Rahimi, M., Moradi, H., Vahabie, A.-h., e Kebriaei, H. (2023). Continuous reinforcement learning-based dynamic difficulty adjustment in a visual working memory game.
Segundo, C., Calixto, K., e Gusmão, R. (2016). Dynamic difficulty adjustment through parameter manipulation for space shooter game. In 2016 15th SBGames, pages 234–237, Brazil. SBC.
Seyderhelm, A. J. A. e Blackmore, K. (2021). Systematic review of dynamic difficulty adaption for serious games: The importance of diverse approaches. SSRN Electronic Journal, 1(1):1–45.
Souza, C., Berreta, L., e de Carvalho, S. (2023a). Performance evaluation of a framework for dynamic difficulty adjustment in single player games (in potuguese). In Proceedings of XI Escola Regional de Informática de Goiás, pages 1–10, Porto Alegre, RS, Brasil. SBC.
Souza, C., Oliveira, S., Berretta, L., e Carvalho, S. (2023b). The use of health data for dynamic difficulty adjustment in serious games (in portuguese). In Proceedings of SBCAS 2023, pages 479–484, Porto Alegre, RS, Brasil. SBC.
Souza, C. H. R., De Oliveira, S. S., Berretta, L. O., e de Carvalho, S. T. (2024). DDAMAPEKit: A framework for dynamic difficulty adjustment based on MAPE-K loop. In Proceedings of the 22nd Brazilian Symposium on Games and Digital Entertainment, SBGames’23, page 1–10, New York, NY, USA. Association for Computing Machinery.
Streicher, A. e Smeddinck, J. D. (2016). Personalized and adaptive serious games. In Entertainment Computing and Serious Games, pages 332–377. Springer International Publishing, USA.
Tagliaro, L. R. G. (2022). An implementation of adaptive difficulty systems for challenging video games. Undergraduate Thesis.
Wei, J., Wang, X., Schuurmans, D., Bosma, M., Ichter, B., Xia, F., Chi, E., Le, Q., e Zhou, D. (2022). Chain-of-thought prompting elicits reasoning in large language models.
Weyns, D. (2021). An Introduction to Self-Adaptive Systems. Wiley, USA.
Zohaib, M. (2018). Dynamic difficulty adjustment (DDA) in computer games: A review. Advances in Human-Computer Interaction, 2018:1–12.
Figueira, F., Nascimento, L., Junior, J., Kohwalter, T., Murta, L., e Clua, E. (2018). BinG: A framework for dynamic game balancing using provenance. In 2018 17th SBGames, pages 57 – 66, Brazil. IEEE.
Gallotta, R., Todd, G., Zammit, M., Earle, S., Liapis, A., Togelius, J., e Yannakakis, G. N. (2024). Large language models and games: A survey and roadmap.
Hendrix, M., Bellamy-Wood, T., McKay, S., Bloom, V., e Dunwell, I. (2019). Implementing adaptive game difficulty balancing in serious games. IEEE Transactions on Games, 11(4):320–327.
Hu, S., Huang, T., Ilhan, F., Tekin, S., Liu, G., Kompella, R., e Liu, L. (2024). A survey on large language model-based game agents.
Mi, Q. e Gao, T. (2022). Improved belgian AI algorithm for dynamic management in action role-playing games. Applied Sciences, 12(22):11860.
Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., e Sutskever, I. (2019). Language models are unsupervised multitask learners.
Rahimi, M., Moradi, H., Vahabie, A.-h., e Kebriaei, H. (2023). Continuous reinforcement learning-based dynamic difficulty adjustment in a visual working memory game.
Segundo, C., Calixto, K., e Gusmão, R. (2016). Dynamic difficulty adjustment through parameter manipulation for space shooter game. In 2016 15th SBGames, pages 234–237, Brazil. SBC.
Seyderhelm, A. J. A. e Blackmore, K. (2021). Systematic review of dynamic difficulty adaption for serious games: The importance of diverse approaches. SSRN Electronic Journal, 1(1):1–45.
Souza, C., Berreta, L., e de Carvalho, S. (2023a). Performance evaluation of a framework for dynamic difficulty adjustment in single player games (in potuguese). In Proceedings of XI Escola Regional de Informática de Goiás, pages 1–10, Porto Alegre, RS, Brasil. SBC.
Souza, C., Oliveira, S., Berretta, L., e Carvalho, S. (2023b). The use of health data for dynamic difficulty adjustment in serious games (in portuguese). In Proceedings of SBCAS 2023, pages 479–484, Porto Alegre, RS, Brasil. SBC.
Souza, C. H. R., De Oliveira, S. S., Berretta, L. O., e de Carvalho, S. T. (2024). DDAMAPEKit: A framework for dynamic difficulty adjustment based on MAPE-K loop. In Proceedings of the 22nd Brazilian Symposium on Games and Digital Entertainment, SBGames’23, page 1–10, New York, NY, USA. Association for Computing Machinery.
Streicher, A. e Smeddinck, J. D. (2016). Personalized and adaptive serious games. In Entertainment Computing and Serious Games, pages 332–377. Springer International Publishing, USA.
Tagliaro, L. R. G. (2022). An implementation of adaptive difficulty systems for challenging video games. Undergraduate Thesis.
Wei, J., Wang, X., Schuurmans, D., Bosma, M., Ichter, B., Xia, F., Chi, E., Le, Q., e Zhou, D. (2022). Chain-of-thought prompting elicits reasoning in large language models.
Weyns, D. (2021). An Introduction to Self-Adaptive Systems. Wiley, USA.
Zohaib, M. (2018). Dynamic difficulty adjustment (DDA) in computer games: A review. Advances in Human-Computer Interaction, 2018:1–12.
Publicado
30/09/2024
Como Citar
SOUZA, Carlos H. R.; OLIVEIRA, Saulo S.; BERRETTA, Luciana O.; CARVALHO, Sérgio T..
Large Language Models and Dynamic Difficulty Adjustment: An Integration Perspective. In: TRILHA DE COMPUTAÇÃO – ARTIGOS CURTOS - SIMPÓSIO BRASILEIRO DE JOGOS E ENTRETENIMENTO DIGITAL (SBGAMES) , 2024
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2024
.
p. 31-36.
DOI: https://doi.org/10.5753/sbgames_estendido.2024.241217.
