ARTIFICIAL INTELLIGENCE: BUILDING A BETTER MARIO, WITH SCIENCE
I suppose this might be a bit of a tease bringing it up this close to the deadline, but I’ve only just spotted Julian Togelius and Sergey Karakovskiy’s Mario AI Competition 2009, a contest in association with the IEEE, to build an algorithm capable of controlling Mario in a modified, endless, randomly generated Java version of Super Mario World.
Explain the coordinators:
One of the main purposes of this competition is to be able to compare different controller development methodologies against each other, both those based on learning techniques such as artificial evolution and those that are completely hand-coded. So we hope to get submissions based on evolutionary neural networks, genetic programming, fuzzy logic, temporal difference learning, human ingenuity, hybrids of the above, etc. The more the merrier! (And better for science.)
There’s two upcoming deadlines, August 18 and September 3, so it may be a bit more feasible to hit that second: check the official site for the full details and source code to enter yourself, and please, please let us know if and when you create something.
See more posts about: Offworld Originals
