Re: [Bug-gnubg] Forwarded suggestion from the website

[Øystein Johansen]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jim Segrave wrote:
| I thought Joseph said they only come into play on .gt. 0 ply rollouts
| (they speed up the selection of candidates for the next ply), so
| they'd have no effect on zero ply rollouts.

Yes, they are. But the move selected in the rollout will be done with
the normal nets. It's the luck connected to the each roll that will be
calculated with the pruning nets. The luck for each roll is used in the
luck adjusted result used in the rollout.

In a variance reduced rollout it isn't important how how good the luck
is quantified. The most important thing is that it is quatified. I guess
the prune nets can be great for that.

I think it was David Montgomery who wrote a really good article on
variance reduction at GammOnLine. Search for the article there.
Ahaaa: http://www.gammonline.com/members/Feb00/articles/var.htm

There was also someone who suggested that variance reduction in races
can be done really simple, by just assigning a luck based on the dice roll.

- -Øystein
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFBfs/r6kDTFPhwyqYRAsNFAJ9mq95YB6VRzDSTQ1L3+fRdJEyYuQCeKc3M
nAPe3of/QlF7UL/Lt88rdSI=
=lXRI
-----END PGP SIGNATURE-----