[ExI] alpha zero

Dave Sill sparge at gmail.com
Fri Dec 8 16:26:39 UTC 2017


On Fri, Dec 8, 2017 at 10:22 AM, spike <spike66 at att.net> wrote:

>
> ...assuming they didn't do something to StockFish to artificially weaken
> it.  I don't have StockFish, so I can't determine that question myself.
>

Why not? It's free.

I expect this will be proven true or false in the next couple weeks with
> first reports showing up anytime now.  StockFish owners will take those
> games and see if SF really does what the paper reports, given these
> settings.
>
> I can't shake the suspicion that something is amiss.  Evidence to follow.
>

This is from the Stockfish Wikipedia entry, emphasis mine:

  *Stockfish versus AlphaZero[edit]*
*  In December 2017, AlphaZero (developed by Google's DeepMind division)
defeated Stockfish 8 after just 4 hours of self-play, with no access to
opening books or endgame tables, but with superior computing power
allocated to AlphaZero.[26][27][28] Each program was given one minute's
worth of thinking time per move. In 100 games from the normal starting
position AlphaZero won 25 games as White, won 3 as Black, and tied the
remaining 72.[29] In a series of twelve 100-game matches against Stockfish
starting from popular openings, Alphazero won 290, drew 886 and lost
24.[citation needed] The results of the tournaments do not indicate that
Alpha-Go is a superior chess engine - the Alpha-Go algorithm was allocated
more computing power for the games.[27] Google said they are not commenting
on the research until it is published in a journal.[29] The research also
has not yet been peer reviewed.[29]*

-Dave
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20171208/ce523e1f/attachment.html>


More information about the extropy-chat mailing list