Re: On game space size and Monte Carlo sampling

[Mousheng Xu <xu@xxxxxxxxxx>]

At 03:47 PM 4/28/99 -0400, John Aspinall wrote:
>There are two questions here:
> (1) How can I get unbiased samples from my Monte Carlo simulation?
> (2) How many unbiased samples do I need?

>Question 1 is the hard one.  The danger is that the way you initialize your
>Monte Carlo sampler will bias the samples.  The best answer is to use Propp

John,

	If you only care about how much you can save from an exhausted,
brute-force search tree, you don't care the "bias" from real games, you can
assume uniform because it is uniformly distributed (one config, one branch).

-- Mousheng Xu