TD-Gammon, a self-teaching backgammon program, achieves master-level play
Neural Computation
Co-Evolution in the Successful Learning of Backgammon Strategy
Machine Learning
Introduction to Reinforcement Learning
Introduction to Reinforcement Learning
Achieving Efficient and Cognitively Plausible Learning in Backgammon
ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
Challenge-Sensitive Action Selection: an Application to Game Balancing
IAT '05 Proceedings of the IEEE/WIC/ACM International Conference on Intelligent Agent Technology
The case for dynamic difficulty adjustment in games
Proceedings of the 2005 ACM SIGCHI International Conference on Advances in computer entertainment technology
A Theory of Fun for Game Design
A Theory of Fun for Game Design
IJCAI'05 Proceedings of the 19th international joint conference on Artificial intelligence
The Artificial Intelligence
Towards capturing and enhancing entertainment in computer games
SETN'06 Proceedings of the 4th Helenic conference on Advances in Artificial Intelligence
Hi-index | 0.00 |
For video games, matching the game difficulty to the skill level of the player is important for the entertainment, enjoyment and growth of the player. For computer "board games", varying the difficulty has traditionally been left up to the player by providing a fixed number of static difficulty settings. Static skill levels can be problematic for the player: if too many levels are provided, then it can take the user many games to identify the optimal level; if too few levels are provided, then the user can get stuck in between a level that is too easy and a level that is too difficult. In this paper, we examine an approach for dynamically varying the difficulty of the computer opponent during gameplay to better match the skill level of the human player. We provide results for the game of backgammon from tens of thousands of simulated games and show that our dynamic approach better matches the skill level of the opponent than a static approach as measured by a number of metrics.