Neural Content Generation
Machine learning is a hot topic and finds its use in many fields. This research is an attempt to show the possibility to use machine learning to create content for games. Game content is often made using procedural content generation (PCG). PCG has many uses in different games. But this research focuses on PCG for creating worlds, as is done in Minecraft (Fingas J.), Space Engine, Infinity: Battlescape and many other games.
As a proof of concept the goal is to use a neural network to create grayscale heightmaps.