Procedural Game Map Generation using Multi-leveled Cellular Automata by Machine learning
Zhixuan Wu, Yuwei Mao, Qiyu Li · 2021
The concept of Procedural Content Generation (PCG) has been intensively applied in the game industry for its capability of producing infinite game maps without any human effort. Innumerable games, such as Minecraft, Terraria, and No Man's Sky, successfully employed this technique to create unpredictable yet playful gaming experiences. While randomness is essential to adding engaging elements to a game, complete randomness may hurt the outcome of map generations by making a chaotic scene. To address this issue, this paper introduces an effective way of "tweaking" the randomness to generate flexible, endless, natural-looking game maps by machine learning.