It seems that another logical improvement would be to increase the pattern size, but this is not mentioned in the paper. What are the implications of that? Does it result in an imposibly large optimization problem?
Yes, it is maybe a very good idea. I haven't try larger patterns with simulation balancing, but only 3x3 patterns. Maybe larger pattern shape will help.
Aja
_______________________________________________ Computer-go mailing list [email protected] http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
