Towards the automatic invention of simple mixed reality games

The invention of mixed reality games that combine virtual and physical play offers a rich and challenging application area for AI techniques. We look at the possibility of using descriptive machine learning to automatically invent simple mixed reality games. Specifically, we demonstrate that the HR learning system can generate coherent domain knowledge from the noisy play data gathered from a number of simple physical games. We describe how this could be used to support mixed reality game invention, and discuss the prospects for further work in this area.

[1]  Toby Walsh,et al.  On the notion of interestingness in automated mathematical discovery , 2000, Int. J. Hum. Comput. Stud..

[2]  Stephen Muggleton,et al.  Mathematical applications of inductive logic programming , 2006, Machine Learning.

[3]  Bruce H. Thomas,et al.  First Person Indoor/Outdoor Augmented Reality Application: ARQuake , 2002, Personal and Ubiquitous Computing.

[4]  Simon Colton,et al.  Automatic Invention of Fitness Functions with Application to Scene Generation , 2008, EvoWorkshops.

[5]  Simon Colton,et al.  Predictive and Descriptive Approaches to Learning Game Rules from Vision Data , 2006, IBERAMIA-SBIA.

[6]  Mark Billinghurst,et al.  Face to face collaborative AR on mobile phones , 2005, Fourth IEEE and ACM International Symposium on Mixed and Augmented Reality (ISMAR'05).

[7]  Simon Colton,et al.  Automatic Generation of Implied Constraints , 2006, ECAI.

[8]  Hannu Toivonen,et al.  Discovery of frequent DATALOG patterns , 1999, Data Mining and Knowledge Discovery.

[9]  Simon Colton,et al.  Automated conjecture making in number theory using HR, Otter and Maple , 2005, J. Symb. Comput..

[10]  Simon Colton,et al.  Automated Theory Formation in Pure Mathematics , 2002, Distinguished dissertations.

[11]  Michael R. Genesereth,et al.  General Game Playing: Overview of the AAAI Competition , 2005, AI Mag..

[12]  Luc De Raedt,et al.  Clausal Discovery , 1997, Machine Learning.