[computer-go] Positions illustrative of computer stupidity ?

[Chrilly]

----- Original Message ----- 
From: <sylvain.gelly at m4x.org>
To: <computer-go at computer-go.org>
Sent: Thursday, November 23, 2006 1:24 PM
Subject: Re: [computer-go] Positions illustrative of computer stupidity ?


>> I think they will play very strong. Sofar all my tests indicates nice
>> scaling, but I admit I have not tried a proper experiment for a long time
>> since I do not have any extra hardware. Perhaps the Mogo team could do
>>something but the problem is that Mogo is so strong it would beat most
>>programs 100% with modest increases in computation time on 9x9.
>
> What we can say from experiments is that the scaling with time is very 
> good
> with few simulations, but becomes less interesting with a lot of 
> simulations.

This is typical for statistical sampling. The variance of the sample mean is 
a function of the square-root of the sampling size. E.g. in opionion polls 
one typically asks around 1000 people. Increasing the sample-size to e.g. 
10.000 does not really pay-off.
It is more important to have no bias in the selection of the sample than to 
increase the sample-size. And it is also more important to get an unbiased 
answer. E.g. in Austria in all election opinion polls in the last years the 
forecasts for the right-wing Freedom party was always much lower than the 
results at the elections, and for the "Greens" it was the other way round. 
The Greens are hip, the Freedom party has a bad reputation, so the 
interviewed people give a socially accepted answer and make their cross in 
the election box for the Freedom party.

In case of UCT the sample-bias is on purpose. I do not know if this 
accelerates convergence or introduces a systematic bias. A better 
performance on a finite sample sizes does not mean that the sampling 
converges in the limit to the true value. I also do not know if the result 
measured at the end-position has some bias.

Chrilly