Developments on Monte Carlo Go

We have developed two go programs, Olga and Oleg, using a Monte Carlo approach, simpler than Bruegmann’s [Bruegmann, 1993], and based on [Abramson, 1990]. We have set up experiments to assess ideas such as progressive pruning, transpositions, temperature, simulated annealing and depth-two tree search within the Monte Carlo framework. We have shown that progressive pruning alone gives better results than Bruegmann’s Monte Carlo go, which uses transpositions, temperature and simulated annealing. Nevertheless, transpositions and temperature are good speed up enhancements that do not lower the level of the program too much. Moreover, the results of our Monte Carlo programs against knowledgebased programs on 9x9 boards and the ever-increasing power of computers lead us to think that Monte Carlo approaches are worth considering for computer go in the future.