Probabilistic foundations for procedural level generation

Abstract

Procedural content generation (PCG) has become a popular research topic in recent years, but not much work has been done in terms of generalized content generators, that is, methods that can generate content for a wide variety of games without requiring hand-tuning. Probabilistic approaches are a promising avenue for creating more general content generators, and specificially map generators. I am interested in exploring probabilistic techniques that could lead to generalized procedural level generators.