A Design Pattern Approach for Multi-Game Level Generation

Abstract

Existing approaches to multi-game level generation rely upon level structure to emerge organically via level fitness. In this paper, we present a method for generating levels for games in the GVGAI framework using a design pattern-based approach, where design patterns are derived from an analysis of the existing corpus of GVGAI game levels. We created two new generators: one constructive, and one search-based, and compared them to a prior existing search-based generator. Results show that our generator is comparable, even preferred, over the prior generator, especially among players with existing game experience. Our search-based generator also outperforms our constructive generator in terms of player preference.

Procedural Content Generation via Knowledge Transformation (PCG-KT)

“…13: Constructive (left) and search-based (right) generators built from the generalized 3x3s. Reproduced with permission from [38]. something new.…”

“…3) Design Pattern Approach for Multi-Game Level Generation: In [38], Beaupre et al took a generalization and selection approach to procedurally generate new levels of particular games by generalizing across level design patterns from multiple games. This led to new game levels that did not fit within the set of existing levels, such as Pacman levels with collectibles outside of the default Pacman pellets.…”

“…Design Patterns [38] Case-based Reasoning Levels Data Mining Association [39] Apriori Algorithm Game Rules…”

Arquitetura e desenvolvimento de um sistema de geração procedural de múltiplos conteúdos para jogos eletrônicos em tempo real

“…Design Patterns are a more low-level view of the system than architectures, as they are common usages of the creation of specific classes and making them communicate between themselves in known problems, especially in the object-oriented paradigm. We advise the reader to not mistake software design patterns for game design patterns, as the ones presented in the work of (BEAUPRE et al, 2018). They extracted a "design" design pattern after analyzing the corpus levels from the general video game AI framework and proposed a method for generating levels based on said patterns.…”