Hello InGo,

Can you keep it running, to get even more stable results from more

I will let run a few more days.

From how many games are your ratings on CGOS?

From about 200 games.

A question of sports: Could you design (in some fair way) a version

I think it it difficult.

Regards,
Hiroshi Yamashita



----- Original Message ----- From: ""Ingo Althöfer"" <[email protected]>
To: <[email protected]>
Sent: Tuesday, February 14, 2012 4:56 AM
Subject: Re: [Computer-go] How well does pure monte carlo play?


Hello Hiroshi,

very interesting results.


I think, so far pureMcUct played only a few hundred rated games on KGS
Can you keep it running, to get even more stable results from more
than 1,000 rated games?

From how many games are your ratings on CGOS?

Having intransitivities between ratings on different servers is normal.
However, the discrepancy for pureMcUct is really amazing.

A question of sports: Could you design (in some fair way) a version
of Aya where the discrepancy between CGOS- and KGS-performance is larger
than 10 levels (or 1,000 rating points)?

Cheers, Ingo.

-------- Original-Nachricht --------
Datum: Tue, 14 Feb 2012 01:30:12 +0900
Von: "Hiroshi Yamashita" <[email protected]>
An: [email protected]
Betreff: Re: [Computer-go] How well does pure monte carlo play?

Hi,

I tried pure MC + UCT + RAVE 10000 playouts on KGS.
It got 15k. (name is pureMcUct, but it uses Rave.)
http://www.gokgs.com/gameArchives.jsp?user=pureMcUct


Program         playouts  KGS  CGOS(19x19) CGOS(9x9)

GNU Go          classic   6k   1800        1800
AyaMC               600   6k   1741        1992
AyaBot          classic   8k   1522        1750
pureMcUctRave     10000  15k   1380        1595
pureMcUct         10000        1081        1389

AyaMC         is my MC program.
AyaBot        is my old non-MC program.
pureMcUctRave is pure MC + UCT + RAVE.
pureMcUct     is pure MC + UCT.

I was surprised there is big difference KGS rank and CGOS 19x19 rating.
AyaBot and pureMcUctRave's Elo difference is only 142(1522-1380), but
 KGS rank difference is 7 rank (15k-8k).

Regards,
Hiroshi Yamashita


----- Original Message ----- From: "Matthew Woodcraft" <[email protected]>
To: <[email protected]>
Sent: Sunday, January 22, 2012 9:45 AM
Subject: Re: [Computer-go] How well does pure monte carlo play?


> Hiroshi Yamashita wrote:
>> If CGOS rank is right, myCtest-10k-UCT is around 8k on KGS.
>> But I think pure MC+UCT is not so strong.
>> CGOS 19x19 population is few, so it is not reliable.
>> For example, RandomGo's rate 505 is maybe overrated.
>
> Ah, yes. Looking at the cross-table for myCtest-10k-UCT, it has 104/104
> against myCtest-10k, 0/144 against Gnugo level 8ish, and 3/34 against
> intermediate opponents.
>
> I don't think there's enough information for any kind of ranking there.
>
> -M-
> _______________________________________________
> Computer-go mailing list
> [email protected]
> http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
_______________________________________________
Computer-go mailing list
[email protected]
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go

--
Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir
belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de
_______________________________________________
Computer-go mailing list
[email protected]
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go

_______________________________________________
Computer-go mailing list
[email protected]
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go

Reply via email to