Game Level Blending using a Learned Level Representation

TL;DR

This paper introduces Clustering-based Tile Embeddings (CTE), a novel learned level representation method that enables game level blending across unannotated games, outperforming traditional human-annotated approaches.

Contribution

The paper presents CTE, a new learned representation technique that unifies level features across games without human annotation, facilitating more flexible game level blending.

Findings

CTE performs comparably or better than human-annotated representations in blending tasks.

CTE enables blending across unannotated games without additional annotation effort.

The approach is validated on Nintendo games Lode Runner and Zelda.

Abstract

Game level blending via machine learning, the process of combining features of game levels to create unique and novel game levels using Procedural Content Generation via Machine Learning (PCGML) techniques, has gained increasing popularity in recent years. However, many existing techniques rely on human-annotated level representations, which limits game level blending to a limited number of annotated games. Even with annotated games, researchers often need to author an additional shared representation to make blending possible. In this paper, we present a novel approach to game level blending that employs Clustering-based Tile Embeddings (CTE), a learned level representation technique that can serve as a level representation for unannotated games and a unified level representation across games without the need for human annotation. CTE represents game level tiles as a continuous vector representation, unifying their visual, contextual, and behavioral information. We apply this approach to two classic Nintendo games, Lode Runner and The Legend of Zelda. We run an evaluation comparing the CTE representation to a common, human-annotated representation in the blending task and find that CTE has comparable or better performance without the need for human annotation.