> one of the basic problems of go newbies > is their tendency to place the next stone > near to the latest stone of the opponent. > Sometimes this is called the "2-inch heuristic > of beginners".
How about going the other way, forcing Monte-Carlo simulations onto a coarser grid in the hope of quickly finding information that might be lost in full-grid simulations (by analogy with multigrid methods - there even exists something called multigrid Monte-Carlo; perhaps some of the mathematically inclined can have a look?). As a trivial idea, if both playout-parties are forced to play on all odd intersections before being allowed onto the remaining ones (staking broader claims before entering detailed fights), how would that affect the statistics, and is there a benefit to be had from combining coarse and fine simulations? Claus _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
