Toward Co-creative Dungeon Generation via Transfer Learning

Abstract

Co-creative Procedural Content Generation via Machine Learning (PCGML) refers to systems where a PCGML agent and a human work together to produce output content. One of the limitations of co-creative PCGML is that it requires co-creative training data for a PCGML agent to learn to interact with humans. However, acquiring this data is a difficult and time-consuming process. In this work, we propose approximating human-AI interaction data and employing transfer learning to adapt learned co-creative knowledge from one game to a different game. We explore this approach for co-creative Zelda dungeon room generation.

7 pages, 6 figures, Workshop on Procedural Content Generation

References in corpus (3)

• Friend, Collaborator, Student, Manager: How Design of an AI-Driven Game Level Editor Affects Creators
• The VGLC: The Video Game Level Corpus
• An Interaction Framework for Studying Co-Creative AI

Cited by in corpus (1)

• Planar or Spatial: Exploring Design Aspects and Challenges for Presentations in Virtual Reality with No-coding Interface