Determining Game Quality Through UCT Tree Shape Analysis

Upper Confidence Bounds for Trees (UCT) is an important new algorithm for tree searching that has shown itself to be very powerful. For example, in 2009, the Fuego software beat strong professional and amateur Go players by incorporating UCT and pattern recognition a feat which was previously thought to remain decades away. UCT has given rise to new fields of research such as Monte Carlo Tree Search. The hypothesis of this project is that there is a machine learnable relationship between the shape of UCT search trees for a variety of games their quality – how engaging the games are to humans. A statistically significant positive result was identified, representing a machine learning system that may be of great use when helping to classify automatically-generated board games in the future.

[1]  Dean Pomerleau,et al.  ALVINN, an autonomous land vehicle in a neural network , 2015 .

[2]  Constantin F. Aliferis,et al.  An evaluation of machine-learning methods for predicting pneumonia mortality , 1997, Artif. Intell. Medicine.

[3]  David Sidney Parlett,et al.  The Oxford history of board games , 1999 .

[4]  Peter Auer,et al.  Finite-time Analysis of the Multiarmed Bandit Problem , 2002, Machine Learning.

[5]  Yishay Mansour,et al.  A Sparse Sampling Algorithm for Near-Optimal Planning in Large Markov Decision Processes , 1999, Machine Learning.

[6]  Csaba Szepesvári,et al.  Bandit Based Monte-Carlo Planning , 2006, ECML.

[7]  Yngvi Björnsson,et al.  CadiaPlayer: A Simulation-Based General Game Player , 2009, IEEE Transactions on Computational Intelligence and AI in Games.

[8]  David Weimer.  Bibliography , 2018, Medical History. Supplement.

[9]  Ingo Althöfer,et al.  Automatic Generation and Evaluation of Recombination Games , 2010, J. Int. Comput. Games Assoc..

[10]  Martin Müller,et al.  Fuego—An Open-Source Framework for Board Games and Go Engine Based on Monte Carlo Tree Search , 2010, IEEE Transactions on Computational Intelligence and AI in Games.