NYU Tandon

Learning Controllable 3D Level Generators

Procedural content generation via reinforcement learning offers a robust framework for creating complex, traversable 3D environments within voxel-based systems like Minecraft.

• Reinforcement learning combined with neural cellular automata enables the procedural generation of functional 3D voxel environments that satisfy complex constraints like gravity and pathfinding.
• Generative agents successfully master the construction of intricate, stair-based geometries to manipulate path lengths and define spatial boundaries within levels.
• Current models struggle with reliability in large-scale generation, specifically regarding the connection of vertically stacked portals and maintaining consistency over long path sequences.