TY - GEN
T1 - Learning Controllable 3D Level Generators
AU - Jiang, Zehua
AU - Earle, Sam
AU - Green, Michael
AU - Togelius, Julian
N1 - Publisher Copyright:
© 2022 ACM.
PY - 2022/9/5
Y1 - 2022/9/5
N2 - Procedural Content Generation via Reinforcement Learning (PCGRL) foregoes the need for large human-authored data-sets and allows agents to train explicitly on functional constraints, using computable, user-defined measures of quality instead of target output. We explore the application of PCGRL to 3D domains, in which content-generation tasks naturally have greater complexity and potential pertinence to real-world applications. Here, we introduce several PCGRL tasks for the 3D domain, Minecraft. These tasks will challenge RL-based generators using affordances often found in 3D environments, such as jumping, multiple dimensional movement, and gravity. We train agents to optimize each of these tasks to explore the capabilities of existing in PCGRL. The agents are able to generate relatively complex and diverse levels, and generalize to random initial states and control targets. Controllability tests in the presented tasks demonstrate their utility to analyze success and failure for 3D generators. We argue that these generators could serve both as co-creative tools for game designers, and as pre-trained environment generators in curriculum learning for player agents.
AB - Procedural Content Generation via Reinforcement Learning (PCGRL) foregoes the need for large human-authored data-sets and allows agents to train explicitly on functional constraints, using computable, user-defined measures of quality instead of target output. We explore the application of PCGRL to 3D domains, in which content-generation tasks naturally have greater complexity and potential pertinence to real-world applications. Here, we introduce several PCGRL tasks for the 3D domain, Minecraft. These tasks will challenge RL-based generators using affordances often found in 3D environments, such as jumping, multiple dimensional movement, and gravity. We train agents to optimize each of these tasks to explore the capabilities of existing in PCGRL. The agents are able to generate relatively complex and diverse levels, and generalize to random initial states and control targets. Controllability tests in the presented tasks demonstrate their utility to analyze success and failure for 3D generators. We argue that these generators could serve both as co-creative tools for game designers, and as pre-trained environment generators in curriculum learning for player agents.
UR - http://www.scopus.com/inward/record.url?scp=85142344508&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85142344508&partnerID=8YFLogxK
U2 - 10.1145/3555858.3563273
DO - 10.1145/3555858.3563273
M3 - Conference contribution
AN - SCOPUS:85142344508
T3 - ACM International Conference Proceeding Series
BT - Proceedings of the 17th International Conference on the Foundations of Digital Games, FDG 2022
A2 - Karpouzis, Kostas
A2 - Gualeni, Stefano
A2 - Fowler, Allan
PB - Association for Computing Machinery
T2 - 17th International Conference on the Foundations of Digital Games, FDG 2022
Y2 - 5 September 2022 through 8 September 2022
ER -