![Sample our Education journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5932/TOC-Subject-Sample-2021-Education.jpg"})
Abstract
Hex is a challenging strategy board game for two players. To enhance students’ progress in acquiring understanding and practical experience with complex machine intelligence and programming concepts we developed the Machine Intelligence Hex (MIHex) project. The associated undergraduate student assignment is about designing and implementing Hex players and evaluating them in an automated tournament of all programs developed by the class. This article surveys educational aspects of the MIHex project. Additionally, fundamental techniques for game programming as well as specific concepts for Hex board evaluation are reviewed. The MIHex game server and possibilities of tournament organisation are described. We summarise and discuss our experiences from running the MIHex project assignment over four consecutive years. The impact on student motivation and learning benefits are evaluated using questionnaires and interviews.
Acknowledgements
The authors are grateful to all friends, supporters, and participants of the MIHex project at the University of Newcastle, in particular to Michael Quinlan, Nathan Lovell, Neil Wright, and all students and tutors of the courses COMP3330/6380 Machine Intelligence in 2001 – 2004. Thanks also to all colleagues who were involved, in particular, to Daniel Le Berre for his interest in the project in 2001 and to Frederic Maire at QUT for many inspiring games and suggestions. We are grateful to the Discipline of Computer Science and Software Engineering at the University of Newcastle for supporting the MIHex project and course development of COMP3330/6380 Machine Intelligence.
Notes
1. The assignment was set as a team assignment in the first year, as optional team or individual assignment in the second and third year, and as an assignment where students work individually in the fourth year we were running the project.
2. It was noted by John Nash in 1949 (cf. Berlecamp et al., Citation2001): if there were a winning strategy for the second player then the first player could (%) make an arbitrary move and then apply (steal) the strategy of the second player. An additional move is never a disadvantage in Hex. Therefore both players would have a winning strategy which would be a contradiction to the Hex theorem.
4. In Australia the academic year has two fourteen-week semesters. Semester one typically starts towards the end of February and runs until the exams start in mid June.
5. A description of Dijkstra's algorithm can be found in any undergraduate textbook covering graph algorithms, such as the algorithmics textbook by Cormen et al. (Citation2001).
6. In the case of two overlapping templates, an intrusion by the opponent into the cells of the overlapping area can threaten both templates simultaneously with the result that the connection between one set of terminal points is lost.