Believability Testing and Bayesian Imitation in Interactive Computer Games
In imitation learning, agents are trained to carry out certain actions by examining a demonstration of the task at hand. Though common in robotics, little work has been done in translating these concepts to computer games. Given that present-day games generally use antiquated AI techniques which can often lead to stilted, mechanical and conspicuously artificial behaviour, it seems likely that approaches based on the imitation of human players may produce agents which convey a more humanlike impression than their traditional counterparts. At the same time, there exists no formal method of quantifying what constitutes a ‘humanlike’ impression; an equivalent of the Turing test is needed, with the requirement that an agent’s appearance and behaviour be capable of deceiving an observer into misidentifying it as human. The aims of this paper are thus threefold; we describe an approach to the imitation of strategic behaviour and motion, propose a formal method of quantifying the degree to which different agents are perceived as ‘humanlike’, and present the results of a series of experiments using these two systems.
KeywordsComputer Game Experience Level Believability Test Turing Test Human Player
Unable to display preview. Download preview PDF.
- 2.Hayes, G., Demiris, J.: A Robot Controller Using Learning by Imitation. In: Proc. of the 2nd Int. Symposium on Intelligent Robotic Systems, pp. 198–204 (1994)Google Scholar
- 3.Schaal, S.: Movement Planning and Imitation by Shaping Nonlinear Attractors. In: 12th Yale Workshop On Adaptive And Learning Systems (2003)Google Scholar
- 5.Laird, J.E., Lent, M.v.: Interactice Computer Games: Human-Level AI’s Killer Application. In: Proc. AAAI, pp. 1171–1178 (2000)Google Scholar
- 6.Fairclough, C., Fagan, M., MacNamee, B., Cunningham, P.: Research Directions for AI in Computer Games. Technical report, Trinity College Dublin (2001)Google Scholar
- 7.Livingstone, D., McGlinchey, S.: What Believability Testing Can Tell Us. In: Proc. Int. Conf. on Computer Games: AI, Design and Education (2004)Google Scholar
- 9.Laird, J.E., Duchi, J.C.: Creating Human-like Synthetic Characters with Multiple Skill-Levels: A Case Study Using the Soar Quakebot. In: Proc AAAI (2000)Google Scholar
- 10.Gorman, B., Fredriksson, M., Humphrys, M.: QASE – An Integrated API for Imitation and General AI Research in Commercial Computer Games. In: Proc. CGAMES Int. Conf. Computer Games, pp. 207–214 (2005)Google Scholar
- 11.Gorman, B., Humphrys, M.: Towards Integration of Strategic Planning and Motion Modelling in Interactive Computer Games. In: Proc. Int. Conf. Computer Game Design & Technology, pp. 92–99 (2005)Google Scholar
- 12.Thurau, C., Paczian, T., Bauckhage, C.: Is Bayesian Imitation Learning the Route to Believable Gamebots? In: Proc. GAME-ON North America, pp. 3–9 (2005)Google Scholar
- 13.Laird, J.E.: Using a Game to Develop Advanced AI. IEEE Computer, 70–75 (2001)Google Scholar
- 14.Rao, R., Shon, A., Meltzoff, A.: A Bayesian Model of Imitation in Infants and Robots. In: Dautenhahn, K., Nehaniv, C. (eds.) Imitation and Social Learning in Robots, Humans, and Animals: Behavioural, Social and Communicative Dimensions. Cambridge University Press, Cambridge (2004)Google Scholar