We show that the HyperPlay technique, which maintains a bag of updatable models for sampling an imperfect-information game, is more efficient than taking random samples of play sequences. Also, we demonstrate that random sampling may become impossible under the practical constraints of a game. We show the HyperPlay sample can become biased and not uniformly distributed across an information set and present a remedy for this bias, showing the impact on game results for biased and unbiased samples. We extrapolate the use of the technique beyond General Game Playing and in particular for enhanced security games with in-game percepts to facilitate a flexible defense response.

The use of the Monte Carlo playouts as an evaluation function has proved to be a viable, general technique for searching intractable game spaces. This facilitate the use of statistical techniques like Monte Carlo Tree Search (MCTS), but is also known to require significant processing overhead. We seek to improve the quality of information extracted from the Monte Carlo playout in three ways. Firstly, by nesting the evaluation function inside another evaluation function; secondly, by measuring and utilising the depth of the playout; and thirdly, by incorporating pruning strategies that eliminate unnecessary searches and avoid traps. Our experimental data, obtained on a variety of two-player games from past General Game Playing (GGP) competitions and others, demonstrate the usefulness of these techniques in a Nested Player when pitted against a standard, optimised UCT player.

In the field of General Game Playing the imperfectinformationgames present a special challenge for researchers.In general the search space is larger, and thelack of information requires a different decision makingtechnique. A simple Monte Carlo sampling using a particlefilter may serve for the simple games, but this soonfails when more complex games are played. The HyperPlaytechnique was one such ”simple” player, soonenhanced to HyperPlay-II capable of handling the mostcomplex of games. However, this technique is very resourcehungry and may not scale well for larger games.We explore the scalability of HyperPlay-II for a varietyof imperfect-information games and test some perfectinformationpruning techniques to see if they will improveefficiency.

General Game Playing is the design of AI systems able to understand the rules of new games and to use such descriptions to play those games effectively. Games with incomplete information have recently been added as anew challenge for general game-playing systems. The only published solutions to this challenge are based on sampling complete information models. In doing so they ground all of the unknown information, thereby making information gathering moves of no value; a well-known criticism of such sampling based systems. We present and analyse a method for escalating reasoning from complete information models to incomplete information models and show how this enables a general game player to correctly value information in incomplete information games. Experimental results demonstrate the success of this technique over standard model sampling.