In message <[EMAIL PROTECTED]>, Richard Brown <[EMAIL PROTECTED]>
writes
Chaslot G (MICC) wrote:
Question for native English speakers: do you think this technique is
best described by “progressive unpruning” or “progressive widening”?
By neither.
Allow me to suggest a third alternative, one
I think grafting would imply attaching an already-existing structure,
as in genetic programming. This is just about expanding the allowable
area into which the tree grows.
Maybe the bonsai folks have a term for this...
Peter Drake
http://www.lclark.edu/~drake/
On May 24, 2007, at 10:56 AM
If you want to go this way, I would use "progressive branching."
Cheers,
David
On 24, May 2007, at 10:56 AM, Richard Brown wrote:
Allow me to suggest a third alternative, one which I believe to be
best,
"progressive grafting".
___
computer-go m
Yes, unpruning sounds like undoing something previously done.
With our trees we can prune and "unprune," but that is not what
is being discussed. It is the branching growth of the tree, not
cutting some lines of play off and then deciding to bring them
back.
Because we are adding nodes for the fi
From: George Dahl <[EMAIL PROTECTED]>
>Does anyone know of any open source Go AI's written in pure python?
http://senseis.xmp.net/?SimpleGo - early versions appear to have been pure
python.
later versions use a mix of python and c for the monte carlo bits where
performance matters.
On Wed, 23 May 2007, David Doshay wrote:
I thought the first MC Go program was Gobble, 1993, by a physics guy
named Bruegmann. The technique was quite different than today. It was
done as a simulated annealing.
That's the first one I heard of ...
Christoph
If I remember right, the original SimpleBot by Aloril was in python. There
was also a spin off called PyBot. Cheating and looking at
http://www.weddslist.com/kgs/names.html, I believe the spin off was by Deren
Dohoda... who also helped with SimpleBot. The two bots probably represented
different
On 5/24/07, Chaslot G (MICC) <[EMAIL PROTECTED]> wrote:
do you think this technique is best
described by "progressive unpruning" or "progressive widening"?
I'd vote for iterative widening. Or has that been rendered unusable by
Cazenave ;-)
As several others have already pointed out 'unpruning'
some tree heuristics good, some tree heuristics bad.
s.
- Original Message
From: Peter Drake <[EMAIL PROTECTED]>
To: computer-go
Sent: Thursday, May 24, 2007 12:53:03 PM
Subject: Re: [computer-go] Progressive unpruning in Mango 19x19
This interesting -- it implies that the pla
Chaslot G (MICC) wrote:
Question for native English speakers: do you think this technique is
best described by “progressive unpruning” or “progressive widening”?
By neither.
Allow me to suggest a third alternative, one which I believe to be best,
"progressive grafting".
Just as a gardener "p
Widening sounds more natural to me.
On 5/24/07, Chaslot G (MICC) <[EMAIL PROTECTED]> wrote:
Dear all,
I did experiments on 19x19 Mango with 25000 simulations per move, against
GnuGo 3.6 level 0.
Without progressive unpruning, Mango wins 2.9% (250 games), and with
progressive unpruning, Ma
unprune isn't a word in english (yet), so it might be more natural to
use "widening".
you can "un" a lot of things, but pruning is generally a somewhat
irreversible action.
s.
- Original Message
From: Brian Slesinsky <[EMAIL PROTECTED]>
To: computer-go
Sent: Thursday, May 24, 2007 12:
Question for native English speakers: do you think this technique is
best described by “progressive unpruning” or “progressive widening”?
I used this term in reference to Tristan Cazenave's "iterative widening"
and "generalized widening" (I should have cited him). See:
http://www.ai.univ-paris
Does anyone know of any open source Go AI's written in pure python?
Thanks,
George
___
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/
Yes, my recent (unsuccessful) experiments have also been along these
lines. It's nice to know I wasn't barking up the wrong tree after all!
Peter Drake
http://www.lclark.edu/~drake/
On May 24, 2007, at 9:35 AM, [EMAIL PROTECTED] wrote:
For the typical person in the U.S. the thing you prun
On 5/24/07, John Tromp <[EMAIL PROTECTED]> wrote:
> Question for native English speakers: do you think this technique is best
> described by "progressive unpruning" or "progressive widening"?
I'm no native speaker, but I think using the word "selectivity" may be
most descriptive.
Does "regressiv
I'd have to read the paper to make sure I understand what's being done, but to
my ears,
"progressive widening" is more natural and descriptive than "progressive
unpruning."
Terry McIntyre
UNIX for hire
software development / systems administration / security
[EMAIL PROTECTED]
- Original M
This interesting -- it implies that the place to use the heuristics
IS in the tree rather than in the playouts.
Peter Drake
http://www.lclark.edu/~drake/
On May 24, 2007, at 8:50 AM, Chaslot G (MICC) wrote:
Dear all,
I did experiments on 19x19 Mango with 25000 simulations per move,
ag
On 5/24/07, Chaslot G (MICC) <[EMAIL PROTECTED]> wrote:
Question for native English speakers: do you think this technique is best
described by "progressive unpruning" or "progressive widening"?
I think "progressive widening" sounds a little better.
- Brian
For the typical person in the U.S. the thing you prune would naturally be a
bush. "Unpruning" something brings up a humorous mental picture of vainly
trying to stick the clipped branches back on.
I should mention that while the idea is not new (and I've tried it
unsuccessfully myself in the pa
"Widening" sounds more natural to me.
Peter Drake
http://www.lclark.edu/~drake/
On May 24, 2007, at 8:50 AM, Chaslot G (MICC) wrote:
Dear all,
I did experiments on 19x19 Mango with 25000 simulations per move,
against GnuGo 3.6 level 0.
Without progressive unpruning, Mango wins 2.9% (2
Quoting Peter Drake <[EMAIL PROTECTED]>:
In previous versions of Orego, I have added one node per playout. I
just changed that to add a child to a node only if that node has at
least A runs, where A is the area of the board (e.g., 81). This seems
to make the program stronger, if only becaus
Question for native English speakers: do you think this technique is best
described by "progressive unpruning" or "progressive widening"?
I'm no native speaker, but I think using the word "selectivity" may be
most descriptive.
Does "regressive selectivity" sound too weird ?
regards,
-John
_
Dear all,
I did experiments on 19x19 Mango with 25000 simulations per move,
against GnuGo 3.6 level 0.
Without progressive unpruning, Mango wins 2.9% (250 games), and with
progressive unpruning, Mango wins 31% (400 games).
I proposed progressive unpruning in this paper:
http://www.cs.unim
> ... It was done as a simulated annealing.
Yep:
http://nngs.ziar.net/cgml/split/7/5/9/[EMAIL PROTECTED]
And
http://www.ideanest.com/vegos/MonteCarloGo.pdf
___
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/lis
> ... It was done as a simulated annealing.
Yep:
http://nngs.ziar.net/cgml/split/7/5/9/[EMAIL PROTECTED]
[ maybe simulated annealing is Monte Carlo as performed by
blacksmiths, after all ;-]
AvK
___
computer-go mailing list
computer-go@computer-go.org
h
> Alternatively, I wonder if there is some theoretical
> way to work it out?
> What is the most extreme example of being behind
> (either by X stones, or
> by some percentage, such as Heikki's 50% above)
I think the bias comes as MCGO needs to "finish the
game up to the last stone/point"... Killin
The first Monte-Carlo Go program was Gobble programmed by Brügmann in 1993.
Scientific American if full of this kind of blunders, but if they are not in
your field you don't see them.
> John, it says the new algorithm can topple strong players - shall we just
> believe them and say I won that b
28 matches
Mail list logo
