Generating levels that teach mechanics

Abstract

The automatic generation of game tutorials is a challenging AI problem. While it is possible to generate annotations and instructions that explain to the player how the game is played, this paper focuses on generating a gameplay experience that introduces the player to a game mechanic. It evolves small levels for the Mario AI Framework that can only be beaten by an agent that knows how to perform specific actions in the game. It uses variations of a perfect A* agent that are limited in various ways, such as not being able to jump high or see enemies, to test how failing to do certain actions can stop the player from beating the level.This section discusses frameworks and research relevant to our work. It starts with a description of the Mario AI framework, followed by a brief background on search based level generation and level generation for the Mario AI framework, and concluding with tutorials and tutorial generation. arXiv:1807.06734v4 [cs.AI] 1 Oct 2018 FDG18, August 7-10, 2018, Malmö, Sweden M. Green et al.

Automatic Generation of Board Game Manuals

“…One improvement on this work could be the creation of tutorial scenarios where a specific mechanic or rule needs to be understood in order to win, thus providing a more interactive learning experience [8]. Interactive tutorials have been shown to increase player engagement and ability, particularly for complex games [1,12].…”

Mario Level Generation From Mechanics Using Scene Stitching

“…Khalifa et al [13] evolved levels for bullet hell games via constrained Map-Elites, a hybrid evolutionary search, using automated playing agents that mimiced different human playstyles. Within the Mario AI Framework, Green et al [7] proposed a method to automatically generate mini-levels in the Mario AI Framework, called "scenes", which focused on requiring the player to trigger a specific mechanic in order to win. By evolving scenes using constrained Map-Elites, Khalifa et al [12] built upon this research and was able to generate a multitude of levels that featured various subsets of game mechanics.…”