Title :
Teaching game AI through Minecraft mods
Author :
Bayliss, Jessica D.
Author_Institution :
Sch. of Interactive Games & Media, Rochester Inst. of Technol., Rochester, NY, USA
Abstract :
One of the issues with teaching artificial intelligence (AI) for games is that many AI algorithms work in theory, but have production consequences in terms of speed or memory when actually used in a game. We report on the benefits and drawbacks of modifying or “modding” the commercial game Minecraft for a course on Game AI. This was done to give students the experience of dealing with a commercial game environment where they would have to worry about production consequences with their algorithms. The course was run as an upper level undergraduate elective during the fall of 2011 and included assignments on dynamic terrain generation, character behavior, and world events.
Keywords :
artificial intelligence; computer games; computer science education; educational courses; further education; teaching; Minecraft mods; artificial intelligence teaching; character behavior; commercial game environment; dynamic terrain generation; game AI course; game AI teaching; undergraduate elective; world events; Algorithm design and analysis; Artificial intelligence; Computer architecture; Games; Planning; Servers; Technological innovation; Minecraft; artificial intelligence; curriculum; game development;
Conference_Titel :
Games Innovation Conference (IGIC), 2012 IEEE International
Conference_Location :
Rochester, NY
Print_ISBN :
978-1-4673-1359-9
DOI :
10.1109/IGIC.2012.6329841