Adapting to Human Gamers Using Coevolution

No matter how good a computer player is, given enough time human players may learn to adapt to the strategy used, and routinely defeat the computer player. A challenging task is to mimic this human ability to adapt, and create a computer player that can adapt to its opposition’s strategy. By having an adaptive strategy for a computer player, the challenge it provides is ongoing. Additionally, a computer player that adapts specifically to an individual human provides a more personal and tailored game play experience. To address this need we have investigated the creation of such a computer player. By creating a computer player that changes its strategy with influence from the human strategy, we have shown that the holy grail of gaming – an individually tailored gaming experience, is indeed possible. We designed the computer player for the game of TEMPO, a zero sum military planning game. The player was created through a process that reverse engineers the human strategy and uses it to coevolve the computer player.

[1]  Sushil J. Louis,et al.  Combining Case-Based Memory with Genetic Algorithm Search for Competent Game AI , 2005, ICCBR Workshops.

[2]  Sushil J. Louis,et al.  Trap Avoidance in Strategic Computer Game Playing with Case Injected Genetic Algorithms , 2004, GECCO.

[3]  Zbigniew Michalewicz,et al.  Coevolving strategic intelligence , 2008, 2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence).

[4]  Paul F. Reynolds,et al.  Exploring the constraints of human behavior representation , 2004, Proceedings of the 2004 Winter Simulation Conference, 2004..

[5]  Steve Rabin,et al.  AI Game Programming Wisdom , 2002 .

[6]  David W. Aha,et al.  Learning to Win: Case-Based Plan Selection in a Real-Time Strategy Game , 2005, Künstliche Intell..

[7]  Susan L. Epstein Game Playing: The Next Moves , 1999, AAAI/IAAI.

[8]  Elizabeth Sklar,et al.  Animal-animat coevolution: using the animal population as fitness function , 1998 .

[9]  Sandra Carberry,et al.  Techniques for Plan Recognition , 2001, User Modeling and User-Adapted Interaction.

[10]  Ian Lane Davis,et al.  Strategies for Strategy Game AI , 1999 .

[11]  Zbigniew Michalewicz,et al.  Coevolutionary TEMPO game , 2004, Proceedings of the 2004 Congress on Evolutionary Computation (IEEE Cat. No.04TH8753).

[12]  Sushil J. Louis,et al.  Playing to learn: case-injected genetic algorithms for learning to play computer games , 2005, IEEE Transactions on Evolutionary Computation.

[13]  Sushil J. Louis,et al.  Learning to play like a human: case injected genetic algorithms for strategic computer gaming , 2004, Proceedings of the 2004 Congress on Evolutionary Computation (IEEE Cat. No.04TH8753).

[14]  Katrin Becker,et al.  Teaching with games: the Minesweeper and Asteroids experience , 2001 .

[15]  Jordan B. Pollack,et al.  Measuring Progress in Coevolutionary Competition , 2000 .

[16]  Pieter Spronck,et al.  Adaptive game AI , 2005 .

[17]  Sushil J. Louis,et al.  Learning with case-injected genetic algorithms , 2004, IEEE Transactions on Evolutionary Computation.

[18]  Eric O. Postma,et al.  Adaptive game AI with dynamic scripting , 2006, Machine Learning.

[19]  Luc Lamontagne,et al.  Case-Based Reasoning Research and Development , 1997, Lecture Notes in Computer Science.

[20]  David W. Aha,et al.  Knowledge acquisition for adaptive game AI , 2007, Sci. Comput. Program..

[21]  David W. Aha,et al.  Automatically Generating Game Tactics through Evolutionary Learning , 2006, AI Mag..