Visualising Multiplayer Game Spaces
This work addresses the problem of analyzing game characteristics for researchers in game AI, but it is incremental as it builds on existing visualization and classification methods.
The study compared four game-space visualizations to characterize multiplayer tabletop games, finding that a space based on Monte Carlo Tree Search (MCTS) optimization was most interpretable for assessing factors like imperfect information and adversarial opponents, but no general effect of player count was observed in dimensionality reduction.
We compare four different `game-spaces' in terms of their usefulness in characterising multi-player tabletop games, with a particular interest in any underlying change to a game's characteristics as the number of players changes. In each case we take a 16-dimensional feature space, and reduce it to a 2-dimensional visualizable landscape. We find that a space obtained from optimization of parameters in Monte Carlo Tree Search (MCTS) is the most directly interpretable to characterise our set of games in terms of the relative importance of imperfect information, adversarial opponents and reward sparsity. These results do not correlate with a space defined using attributes of the game-tree. This dimensionality reduction does not show any general effect as the number of players. We therefore consider the question using the original features to classify the games into two sets; those for which the characteristics of the game changes significantly as the number of players changes, and those for which there is no such effect.