Improving Dynamic Scripting for Adaptive Game AI with a Tactic Replacement Algorithm
Resumo
Artificial Intelligence (AI) plays an important role in digital games nowadays. With players becoming increasingly demanding, it is vital to provide an AI that challenges and entertains them. The use of Adaptive Artificial Intelligence (AAI) has shown potential to adapt to each player by learning their techniques and offering a consistent challenge. This research consists on the analysis of an AAI technique known as Dynamic Scripting (DS) and in the development of a new algorithm (called Tactic Replacement) to improve it. Results show that, in comparison with the default DS algorithm, the proposed algorithm achieved a time reduction of ≈ 50% to achieve convergence. Also, it was able to reduce by 40% the average number of rounds to reach the convergence.
Referências
Armstrong, W., Christen, P., McCreath, E., and Rendell, A. (2006). Dynamic Algorithm selection Using Reinforcement Learning. 2006 International Workshop on Integrating AI and Data Mining, pages 18–25.
Bateman, C. M. (2007). Game writing : narrative skills for videogames. Cengage Learning; 1 edition (July 3, 2006).
Bioware (2000). Neverwinter Nights.
Champandard, A. J. (2003). AI Game Development: Synthetic Creatures with Learning and Reactive Behaviors. NRG Series. New Riders.
Mitchell, T. M. (1997). Machine Learning. McGraw-Hill, Inc., New York, NY, USA, 1 edition.
Mu˜noz-Avila, H., Bauckhage, C., Bida, M., Congdon, C., and Kendall, G. (2013). Learning and Game AI. Artificial and Computational Intelligence in Games, 6:33–43.
Rabin, S. (2002). AI Game Programming Wisdom. Charles River Media; 1 edition (April 3, 2002), Boston, first edit edition.
Rabin, S. (2013). Game AI Pro: Collected Wisdom of Game AI Professionals. Taylor & Francis.
Russell, J.S. e Norvig, P. (2003). Artificial Intelligence: A Modern Approach. Pearson; 3 edition (December 11, 2009).
Shafer, J. (2012). The Recipe for Good AI. Disponível em: http://ubm.io/2j7DXFY. Acesso em: 31 de Outubro de 2016.
Spronck, P. (2005). Adaptive Game AI. PhD thesis.
Spronck, P., Ponsen, M., Sprinkhuizen-Kuyper, I., and Postma, E. (2006). Adaptive game AI with dynamic scripting. Machine Learning, 63(3):217–248.
Spronck, P., Sprinkhuizen-Kuyper, I., and Postma, E. (2004). Difficulty scaling of game AI. Proceedings of the 5th International Conference on Intelligent Games and Simulation (GAMEON’2004), pages 33–37.
Sutton, R. and Barto, A. (1998). Reinforcement Learning: An Introduction. IEEE Transactions on Neural Networks, 9(5):1054–1054.
Thawonmas, R. and Osaka, S. (2006). A method for online adaptation of computergame AI rulebase. International Conference on Advances in Computer Entertainment Technology 2006, page 16.
Wexler, J. (2008). Artificial Intelligence in Games: A look at the smarts behind Lionhead Studio’s “Black and White” and where it can and will go in the future. Spring Simulation Multiconference.
