Tile Embedding: A General Representation for Procedural Level Generation via Machine Learning
This addresses the bottleneck of dataset creation for game developers and researchers, though it is incremental as it builds on existing autoencoder methods.
The paper tackles the problem of limited annotated datasets for Procedural Level Generation via Machine Learning (PLGML) in tile-based 2D games by introducing tile embeddings, a unified representation that reduces the need for human annotation, enabling application to more games beyond common examples like Super Mario Bros.
In recent years, Procedural Level Generation via Machine Learning (PLGML) techniques have been applied to generate game levels with machine learning. These approaches rely on human-annotated representations of game levels. Creating annotated datasets for games requires domain knowledge and is time-consuming. Hence, though a large number of video games exist, annotated datasets are curated only for a small handful. Thus current PLGML techniques have been explored in limited domains, with Super Mario Bros. as the most common example. To address this problem, we present tile embeddings, a unified, affordance-rich representation for tile-based 2D games. To learn this embedding, we employ autoencoders trained on the visual and semantic information of tiles from a set of existing, human-annotated games. We evaluate this representation on its ability to predict affordances for unseen tiles, and to serve as a PLGML representation for annotated and unannotated games.