Quoting Petr Baudis [EMAIL PROTECTED]:
Please note that pachi1 had a rather embarassing bug of starting the
random playouts with wrong color (so if the last tree node was black,
the playout would start with black as well). pachi2 has this bug fixed;
the ELO rating is still not settled, but so
Magnus Persson wrote:
Quoting Petr Baudis [EMAIL PROTECTED]:
Please note that pachi1 had a rather embarassing bug of starting the
random playouts with wrong color (so if the last tree node was black,
the playout would start with black as well). pachi2 has this bug fixed;
the ELO rating is
Petr Baudis wrote:
I got kind of lost in the thread and lost track about which bots should
I actually compare myself to. ;-)
So I have created this page:
http://senseis.xmp.net/?CGOSBasicUCTBots
Good idea.
and summed up what I could find in the thread about the various bots.
Don Dailey wrote:
Tim Foden wrote:
Don Dailey wrote:
I suggest
exactly 25,000 play-outs that we should standardize on.50,000 will
tax my spare computer which I like to use for modest CGOS tests.
If it is agreed, I will start a 25k test.My prediction is that this
will finish
If it is agreed, I will start a 25k test.My prediction is that this
will finish around 1600 ELO on CGOS.
Looks like it's currently around 1485, so I am 115 ELO off from my
prediction at the moment.
2 more doublings would probably get you to 1700 - perhaps I will test
this later,
On Thu, Mar 13, 2008 at 09:19:58AM -0400, Don Dailey wrote:
If it is agreed, I will start a 25k test.My prediction is that this
will finish around 1600 ELO on CGOS.
Looks like it's currently around 1485, so I am 115 ELO off from my
prediction at the moment.
2 more
On Thu, Mar 13, 2008 at 09:19:58AM -0400, Don Dailey wrote:
If you look at the table, drdGeneric 10k is rated 1228 and 25k is
rating 1485 which is 257 ELO for doing 2.5 x more play-outs. If
this holds, I would expect 100,000 play-outs to give well over 1700
ELO. Of course there could
On Thu, 13 Mar 2008, Petr Baudis wrote:
So I have created this page:
http://senseis.xmp.net/?CGOSBasicUCTBots
and summed up what I could find in the thread about the various bots.
Please clarify if anything there is wrong / unknown, and add your bots
if they aren't there. I wanted to
On Thu, Mar 13, 2008 at 11:25:04AM -0700, Christoph Birk wrote:
On Thu, 13 Mar 2008, Petr Baudis wrote:
So I have created this page:
http://senseis.xmp.net/?CGOSBasicUCTBots
and summed up what I could find in the thread about the various bots.
Please clarify if anything there is wrong
On Thu, Mar 13, 2008 at 06:31:19PM +0100, Petr Baudis wrote:
I got kind of lost in the thread and lost track about which bots should
I actually compare myself to. ;-)
So I have created this page:
http://senseis.xmp.net/?CGOSBasicUCTBots
Good idea!
Would it make sense to have a
On Thu, 13 Mar 2008, Heikki Levanto wrote:
Would it make sense to have a similar page for pure MC programs (without
uct), so that we beginning developers could check that portion of our code
against known results?
I have two long-term CGOS programs:
myCtest-10k: 1011 ELO
myCtest-50k: 1343
On Thu, Mar 13, 2008 at 08:01:57PM +0100, Heikki Levanto wrote:
On Thu, Mar 13, 2008 at 06:31:19PM +0100, Petr Baudis wrote:
I got kind of lost in the thread and lost track about which bots should
I actually compare myself to. ;-)
So I have created this page:
On Mar 11, 2008, at 2:41 PM, Christoph Birk [EMAIL PROTECTED] wrote:
On Tue, 11 Mar 2008, Don Dailey wrote:
I am going to keep the 25k playouts running and add a 10k play-out
version of UCT. I want to establish a standard testing size so
that
Great! That way Jason can also
Don Dailey wrote:
I suggest
exactly 25,000 play-outs that we should standardize on.50,000 will
tax my spare computer which I like to use for modest CGOS tests.
If it is agreed, I will start a 25k test.My prediction is that this
will finish around 1600 ELO on CGOS.
OK, I added
Tim Foden wrote:
Don Dailey wrote:
I suggest
exactly 25,000 play-outs that we should standardize on.50,000 will
tax my spare computer which I like to use for modest CGOS tests.
If it is agreed, I will start a 25k test.My prediction is that this
will finish around 1600 ELO on
On Mar 12, 2008, at 2:17 PM, Don Dailey [EMAIL PROTECTED] wrote:
Tim Foden wrote:
Don Dailey wrote:
I suggest
exactly 25,000 play-outs that we should standardize on.50,000
will
tax my spare computer which I like to use for modest CGOS tests.
If it is agreed, I will start a 25k
By the way, I never experimented with the formula I
use, I just plagiarized it from information posted here. I'm working
on my own fomula in the meantime and used to use something different I
made up.
What I used before worked like this:
u = (w + n) / (g + n)
u - is what you maximize,
On Wed, 12 Mar 2008, Don Dailey wrote:
1. My UCT constant is 1.0 - my formula is averageScore + c * sqrt(
(2.0 * log(n)) / (10.0 * m) );
so your contstant is 2/10 = 0.2 inside the sqrt(), which is
equivalent to c=0.44 ?
Christoph
___
computer-go
Quoting Don Dailey [EMAIL PROTECTED]:
When the child nodes are allocated, they are done all at once with
this code - where cc is the number of fully legal child nodes:
In valkyria3 I have supernodes that contains an array of moveinfo
for all possible moves. In the moveinfo I also store
On Tue, Mar 11, 2008 at 09:05:01AM +0100, Magnus Persson wrote:
Quoting Don Dailey [EMAIL PROTECTED]:
When the child nodes are allocated, they are done all at once with
this code - where cc is the number of fully legal child nodes:
In valkyria3 I have supernodes that contains an array of
Quoting Jonas Kahn [EMAIL PROTECTED]:
On Tue, Mar 11, 2008 at 09:05:01AM +0100, Magnus Persson wrote:
Quoting Don Dailey [EMAIL PROTECTED]:
When the child nodes are allocated, they are done all at once with
this code - where cc is the number of fully legal child nodes:
In valkyria3 I have
Don Dailey wrote:
Petr Baudis wrote:
On Mon, Mar 10, 2008 at 06:57:07PM -0400, Don Dailey wrote:
I think you may still have a bug. You should get well over 1700 with
110,000 playouts, even if they are light playouts.
Hmmm... That is going to be some tough
On Tue, 11 Mar 2008, Don Dailey wrote:
If it is agreed, I will start a 25k test.My prediction is that this
will finish around 1600 ELO on CGOS.
I have long term rating for simple random playouts:
myCtest-10k and myCtest-50k.
I keep them active since Sept/2006. Please don't use 25k.
If the speed was lowered to 10k, I'd also participate. One of these
days, I'll speed up my engine...
Sent from my iPhone
On Mar 11, 2008, at 11:18 AM, Don Dailey [EMAIL PROTECTED] wrote:
Don Dailey wrote:
Petr Baudis wrote:
On Mon, Mar 10, 2008 at 06:57:07PM -0400, Don Dailey wrote:
This isn't simple random play-outs.It's monte carlo with UCT tree
search.
Ok, I will use 50k to match your test.It means I probably cannot
run 2 tests on that machine and is why I hoped it would be minimal
resource usage, but since you have already started I will restart my test.
-
On Tue, 11 Mar 2008, Don Dailey wrote:
This isn't simple random play-outs.It's monte carlo with UCT tree
search.
Ok, I will use 50k to match your test.It means I probably cannot
run 2 tests on that machine and is why I hoped it would be minimal
resource usage, but since you have
I am going to keep the 25k playouts running and add a 10k play-out
version of UCT. I want to establish a standard testing size so that
I can watch the evolution of the program and 50k is too much if my
program triples in run time as I introduce very heavy play-outs.(I
don't want to count
On Tue, 11 Mar 2008, Don Dailey wrote:
I am going to keep the 25k playouts running and add a 10k play-out
version of UCT. I want to establish a standard testing size so that
Great! That way Jason can also participate.
myCtest-10k-UCT has a long-term rating of about 1250.
For the 50k
On Mon, Mar 10, 2008 at 10:04:18PM -0400, Don Dailey wrote:
Your method is to allocate 1 node when it's been visited once or twice -
very natural I agree. My method is to allocate all the children at
once, and wait until the parent has been visited some number of times
(currently 100). If
Typically, how many parameters do you have to tune ? Real or two-level ?
I guess I have 10 real valued and 10 binary ones. There are probably a lot
of stuff that are ahrd coded and could be parameterized.
Here I am also completely ignoring playouts that have hundreds of handtuned
On Tue, Mar 11, 2008 at 11:41:41AM -0700, Christoph Birk wrote:
On Tue, 11 Mar 2008, Don Dailey wrote:
I am going to keep the 25k playouts running and add a 10k play-out
version of UCT. I want to establish a standard testing size so that
Great! That way Jason can also participate.
Hi,
On Sat, Mar 08, 2008 at 10:18:34AM +0100, Petr Baudis wrote:
(By the way, pachi1-*-light are UCT bots with completely light
playouts with various UCB1 c values, if anyone wants to use that as
reference. Surprisingly, it seems that my heavy playouts do not make big
difference so far,
On Mon, 10 Mar 2008, Petr Baudis wrote:
With 110k playouts per move and no domain knowledge in the playouts,
the ratings are now:
c=0.2 (pachi1-p0.2-light) ELO 1627 (285 games)
c=1.0 (pachi1-p1.0-light) ELO 1590 (120 games)
c=0.05 (pachi1-p0.05-light)
Petr Baudis wrote:
Hi,
On Sat, Mar 08, 2008 at 10:18:34AM +0100, Petr Baudis wrote:
(By the way, pachi1-*-light are UCT bots with completely light
playouts with various UCB1 c values, if anyone wants to use that as
reference. Surprisingly, it seems that my heavy playouts do not
On Mon, Mar 10, 2008 at 03:40:53PM -0700, Christoph Birk wrote:
On Mon, 10 Mar 2008, Petr Baudis wrote:
With 110k playouts per move and no domain knowledge in the playouts,
the ratings are now:
c=0.2 (pachi1-p0.2-light) ELO 1627 (285 games)
c=1.0 (pachi1-p1.0-light)
On Mon, Mar 10, 2008 at 06:57:07PM -0400, Don Dailey wrote:
I think you may still have a bug. You should get well over 1700 with
110,000 playouts, even if they are light playouts.
Hmmm... That is going to be some tough debugging I suspect.
I'm pretty sure my code is fairly well debugged
Petr Baudis wrote:
On Mon, Mar 10, 2008 at 06:57:07PM -0400, Don Dailey wrote:
I think you may still have a bug. You should get well over 1700 with
110,000 playouts, even if they are light playouts.
Hmmm... That is going to be some tough debugging I suspect.
I'm working on
Hi,
On Mon, Mar 10, 2008 at 08:07:14PM -0400, Don Dailey wrote:
What is the justification of using the parent playout count instead of
the node playout count itself?
I don't know if it makes much difference how this is done, and I don't
know how everybody else is doing it. I
On Tue, 11 Mar 2008, Petr Baudis wrote:
On Mon, Mar 10, 2008 at 06:57:07PM -0400, Don Dailey wrote:
I think you may still have a bug. You should get well over 1700 with
110,000 playouts, even if they are light playouts.
I will run myCtest with 110k-playout, c=0.25 and node creation
after the
On Mon, 10 Mar 2008, Christoph Birk wrote:
On Tue, 11 Mar 2008, Petr Baudis wrote:
On Mon, Mar 10, 2008 at 06:57:07PM -0400, Don Dailey wrote:
I think you may still have a bug. You should get well over 1700 with
110,000 playouts, even if they are light playouts.
I will run myCtest with
On Mon, Mar 10, 2008 at 05:36:14PM -0700, Christoph Birk wrote:
On Mon, 10 Mar 2008, Christoph Birk wrote:
On Tue, 11 Mar 2008, Petr Baudis wrote:
On Mon, Mar 10, 2008 at 06:57:07PM -0400, Don Dailey wrote:
I think you may still have a bug. You should get well over 1700 with
110,000
The key point, good or bad, is that I allocate all the legal children
nodes as a group.I manage my own memory pool and just hand out nodes
sequentially.I don't need to maintain pointers to each child, only a
single pointer to the first child. So instead of needing 4 extra bytes
per node
On Mar 10, 2008, at 8:07 PM, Don Dailey [EMAIL PROTECTED] wrote:
Some programs hash each position and the tree is more abstract, no
pointers just positions leading to other positions by zobrist hash
keys in a hash table.
My scheme probably wastes a lot of space on nodes that are left
43 matches
Mail list logo