Scalable Board Expansion within a General Game System

TL;DR

The thesis tackles the inefficiency of fixed-size boards in boardless games by introducing scalable dynamic board expansion within the Ludii General Game System. It presents two expansion strategies—Contour (perimeter-based) and Zone (local-only)—with mapping and reapplication approaches to update topology, container state, and ownership, and analyzes initial-size choices. Through experiments across square, hexagonal, and triangular tile sets and games like Andantino, Bravalath, and Plotto, the PERI-MAP approach emerges as the most effective, delivering substantial improvements in playout throughput and memory efficiency over the baseline. The work demonstrates that dynamic topology and rule-informed expansion can significantly enhance AI performance and fidelity of boardless game representations, laying groundwork for broader adoption and future extensions such as stacking, mixed shapes, and rule-based expansion.

Abstract

This thesis explores the use of a General Game System (GGS) to support the automatic expansion of game boards in boardless games. Traditional implementations of such games often rely on oversized static boards defined from the start, even though large portions of these boards may never be used during gameplay. This approach leads to unnecessary complexity. To address this issue, this thesis propose a dynamic board expansion mechanism in which the game board grows automatically during play.

Figures (51)

• Figure 2: Example of Carcassonne play in an expanded game state carcasonne
• Figure 3: Example of a board during a play
• Figure 4: Graph of 3x3 with square tiles
• Figure 5: Graph of 3x3 with hexagonal tiles
• Figure 6: Graph of 3x3 with triangular tiles