This is the scenario in which you find yourself when you participate in a Scientific Community Game (SCG). Although the game is competitive there is also collaboration: the community of agents collects knowledge about unchallenged and optimal hypotheses. When a hypothesis is discounted, a detailed reason is given that might lead to suggestions how it can be improved.
SCG is a generic game that can be applied to drive innovation and learning in numerous scientific domains in computer science and other disciplines. In computer science, the hypotheses are about problems and their algorithmic solutions, including analyses of the algorithms regarding resource consumption or quality of the solution produced. Innovation happens by understanding the reasons for reputation loss and trying to discover knowledge that other agents might have, which takes place between tournaments, when teams improve their agents.
We have developed a web-based implementation of SCG using DemeterF. Once a domain X is defined, an Administrator(X) and BasicPlayer(X) is produced. Administrator(X) enforces the SCG rules similar to the rules of a real scientific community, guaranteeing the soundness of the game. BasicPlayer(X) is a minimally intelligent agent that can already communicate with other agents through the administrator.
I will summarize different applications of SCG and show you how you can use a manual version of SCG to enhance your interactive learning experience in your CS and Math courses.
SCG (Specker Challenge Game) is named after Ernst Specker who initiated this work and who will soon celebrate his 90th birthday. More information on SCG is available: SCG Home Page.
Joint work with Ahmed Abdelmeged and Bryan Chadwick