On 4/25/07, Remi Munos [EMAIL PROTECTED] wrote:
When I did sit down and read the paper for real, I saw that both of
those
things were just building up the support for the BAST algorithm. (a) is
to
justify accepting a higher regret (in exchange for better worst case
performance).
well,
I haven't gotten to read through your paper as carefully as I'd like yet,
but I do have a few observations that may be of benefit to other readers on
the list... Mostly observation of assumptions used in the paper.
1. A max tree is used instead of minimax
2. Rewards can be more than just 1 or 0
On 4/12/07, Peter Drake [EMAIL PROTECTED] wrote:
The next US Go Congress will be held July 28 - August 4, 2007, in
Lancaster, Pennsylvania:
http://congress.usgo.org/
Is anyone else planning to attend?
I would, at the very least, like to gather a casual meeting of Go
programmers.
I would
Not having byo yomi because it's tough to code isn't really a good
argument. If we want (non-computer-go) people to take the results
seriously, the game timing should be the same as what people naturally do.
I personally am hesitant to play games with fixed time limits because I
don't trust
wrote:
On Thu, 2007-04-12 at 11:18 -0400, Jason House wrote:
Not having byo yomi because it's tough to code isn't really a good
argument. If we want (non-computer-go) people to take the results
seriously, the game timing should be the same as what people naturally
do. I personally am
As with anything, an efficient serial algorithm (alpha-beta, UCT, etc...)
becomes less efficient when made parallel. I think you can see some
significant improvement with parallel machines, but it may be that you'll
get diminishing returns.
I can think of two parallel approaches:
1. Instruct
Jacques Basaldúa wrote:
Daniel Liu wrote:
An imperfect evaluation has errors. Is the exact value of the error
known? No.
I have an idea on that I will try to explain:
Given any finite combinatorial game where the ending nodes
have two possible values: win/loss, any node has a winning rate
I'm doing a redesign of HouseBot and rewriting it in D instead of C++.
Since there was past interest in the D language on this list, I figured
I'd advertise the switch-over. I'm not at the code writing stage yet
because I'm trying to do a proper (re)design.
If the move to D is enough to
I definitely have a number of gripes with the language, but it's not
enough to stop me from trying it out. I didn't realize changes occurred
so quickly, but I don't think they break old code very frequently. Most
changes appear to be bug fixes or additions to the language.
Here's my list of
What is your policy for receiving/handling feature requests? Is there a
tracker (e.g. sourceforge) or a wiki page (e.g. senseis)? Or do we just
e-mail you directly?
On 3/29/07, Don Dailey [EMAIL PROTECTED] wrote:
I aslo recieved many useful suggestions about feature
improvements or additions
After some trial and error, I got 90
* * * *
*
* * * *
* * * ***
* * *
*** * * *
* * * *
*
* * * *
On 3/29/07, John Tromp [EMAIL PROTECTED] wrote:
On 3/29/07, John Tromp [EMAIL PROTECTED] wrote:
Is 88 the maximum number of pseuoliberties a string can have on 9x9?
Make
Arthur W Cater wrote:
It's really a way to incrementally update liberties in a
fast way - each stone keeps it's own count of liberties
and it is summed - but of course it doesn't represent
the true number of liberties since a point can get
counted 2 or more times.However, if the count goes
Chris Fant wrote:
Once upon a time, I did analysis of the inaccuracy of pseudo liberties.
Searching quickly, I found:
http://computer-go.org/pipermail/computer-go/2005-October/003839.html
For any interested, I did come up with a variant of pseudo liberties
that was a lot closer to real
I don't see any games that have an outcome other than winning by points
or resignation. Any forfeits or games that are on hold?
Don Dailey wrote:
I have a prototype of the new CGOS server up and running.
Please help me test it. I have set up a 2 MINUTE SERVER, i.e.
2 minutes per side for
There are some freely available regression test suites. Two come to mind:
* computer go test collection 2.0
* The regression suite available in gnu go
Also, go gui has a program called gtpregress that can handle both of
those out of the box (I think)
Brian Slesinsky wrote:
I've been lurking
Jacques Basaldúa wrote:
Hello,
Just an explanation on something I may have explained badly. I see we
agree in the fundamental.
Correcting bias in that estimate should lead to better sampling.
This is usually called continuity correction
http://en.wikipedia.org/wiki/Continuity_correction.
alain Baeckeroot wrote:
(I propose to ban the term temperature from CGT, and replace it by value,
unless someone can explain the link with temperature in physics, and shows
some identical properties ;-)
While I bet most of us dislike the term, it seems to express an inherent
concept.
I respond to various items below. Sections of the original e-mail that
I'm not responding to were completely deleted.
Jacques Basaldúa wrote:
Hello Jason
I think what you are trying to do can be done more easily.
I guess the key question is what am I trying to do?.
In UCT, the next move
Based on my analysis, estimating a moves probability of winning by
taking the number of winning simulations (w) and dividing it by the
total number of simulations (n) is actually biased. I tried to break
this e-mail up into sections for easy digestion by the various people
who might read
Heikki Levanto wrote:
To get down to earth, I would like to look at the board at the end of
each playout, calculate something more than just win/loss, and pass that
info back to who ever called playout. One way to do that would be to
pass a function pointer and a (void?) pointer to playout, and
?ukasz Lew wrote:
Generally http://en.wikipedia.org/wiki/Disjoint-set_data_structure
In my program it's a tree for each group of stones. In the root is
number of pseudoliberties.
Joining is cheap and checking group membership is cheap.
Look at chain_t class. in board.cpp
Best,
?ukasz Lew
I
Dmitry Kamenetsky wrote:
I have been reading this list for nearly a year now and it is very discouraging
to receive so much criticism for my first post.
Sadly, you chose a topic that is close to the hearts of many. I like
the style of this follow up post much more than the original in
501 - 522 of 522 matches
Mail list logo