[Computer-go] How well does pure monte carlo play?

[Christoph Birk]

On Jan 22, 2012, at 12:34 AM, Petr Baudis wrote:

> On Sat, Jan 21, 2012 at 03:11:30PM -0800, Christoph Birk wrote:
>> 
>> On Jan 21, 2012, at 2:16 PM, Petr Baudis wrote:
>> 
>>> On Sat, Jan 21, 2012 at 03:42:57PM -0500, Don Dailey wrote:
>>>> On Sat, Jan 21, 2012 at 2:31 PM, Eric Baum <ebaum at fastmail.fm> wrote:
>>>> 
>>>>> If you don't do rave, or use patterns to predict likely moves, or any of
>>>>> that, but just do pure monte carlo,
>>>>> parallelized on a cluster,how well would that play Go? Anyone know?
>>> 
>>> I assume you are talking about 9x9? A rather optimistic estimate would
>>> be 1600 Elo on CGOS for 50k playouts per move; with GNUGo at 1800 Elo,
>>> we could estimate this as 8k KGS.
>> 
>> myCtest-50k does exactly that and it rating is 1689 +- 15 (4000 games).
> 
> With what parameters? Is it really plain UCT? It used to be weaker...

It has always been plain UCT with k=0.5 ( uct_score = k*sqrt(log(n/p))
 The only thing I changed about a year ago is avoiding super-ko.

Christoph