Dan Ross, a student of M.S. in Applied Mathematics program, will give a presentation on his thesis, titled “Defining the Problem of Solving Reservi.”

## Abstract:

In the early part of this millennium, it was predicted that two well-known games would be solved by the end of the decade: checkers and reversi (Herik, Uiterwijk, and Rijswijck 2002, 306). In 2007, the first half of this prediction became a reality when it was announced that, with perfect play, or the strategy which leads to the optimal outcome regardless of the response(s) of his opponent(s), either player in a game of checkers can force a draw. Sorting through the 5 x 1020 possible checkers positions required around-the-clock computing on, at times, up to 200 processors for nearly twenty years (Schaeffer et al. 2007, 1518). It currently appears unlikely that reversi, with an estimated 1028 possible positions and significantly more possible games than checkers (Allis 1994, 167-8), will be solved before the end of 2010.

Solving the game of reversi would be a multi-year effort with inherent extensive computing and storage requirements unavailable to the author of this thesis. It has, however, been said that a well-defined problem is half solved; the intent of this thesis is to define the characteristics of reversi in such a way that it could be solved given the proper resources. Results from the solution of a reduced form of the game show that equivalent positions can be generated at a reasonable processing cost to greatly reduce physical storage requireme

Time: 3:30 p.m. on March 29, 2010

Location: Stright 302