Re: [Computer-go] new bot friendly go server.

[Henry Hemming]

I'll try to address the issues brought up as best as I can.

Regarding the game time messages (opponent resigned, please capture dead
stones, etc) not being easy to notice, I'm working on it. However the
language barrier shouldn't be a problem as the website recognizes locales
of jp, cn, kr and is translated to those languages, albeit Korean
translation is still of rather low quality. As for letting players mark
dead stones, it only causes more problems than it solves. Currently the
algorithm recognizes dead groups correctly 90% of the time, I will improve
on the algorithm (probably train a dcnn to do it). For now the remaining
10% of the time the players unfortunately have to play a few extra moves to
clear up the situation for the server.

Regarding choosing names, its actually paid feature (info in the FAQ),
partly because I have to check that the name isn't offensive. However, in
case of well known bots I can make an exception, they are after all there
partly to get visibility among go players and its only fair they get to use
their own name. Marking players as bots will probably happen in the near
future. As for your game Remi, your opponent resigned, there should have
been both audio and text notifications of it happening (the text appears
where the resign and pass buttons are during the game). I will probably
create some way of watching games in real time, but for now you can find
the links to recently played games in an rss, to which there is a link in
the FAQ.

Any and all feedback is much appreciated.

As for comparison to OGS, it tries to be everything to everyone, which just
isn't my cup of tea. I wanted to create a simple server, with reasonable
settings, where you can just play without having to worry about the
irrelevant details and trust you get good match ups (at least after the
first few games to establish your rating).

My rating system... well I'm going to try to get it published and as such
I'm not yet ready to discuss it publicly. But I can reveal the results I
got, tested it on 32k professional go games, 230k dgs games and 5.4mil
goquest games. The error rates on test data (after optimizing hyper
parameters on separate training data) were 48.254%, 46.679%, 47.072%,
comparable numbers for Elo were 48.321%, 47.795%, 48.845%, TrueSkill were
48.313%, 46.914%, 47.194%, and for Glicko 48.289%, 46.704%, 47.127%.
Runtime is O(1) which is roughtly 50% slower than Glicko or Trueskill. I
hope it serve well on my go server :)

On Mon, May 23, 2016 at 6:02 PM daniel rich  wrote:

> Nice, Thanks Henry!
>
> I am in the process of bringing up my first bot(as are a bunch of people)
> so I will probably try and connect in the next couple of days.
>
> --Danny
>
> On Mon, May 23, 2016 at 8:46 AM, Petr Baudis  wrote:
>
>>   Hi!
>>
>> On Sun, May 22, 2016 at 08:56:09AM +, Henry Hemming wrote:
>> > Hello, I would like to invite all you go bot developers to my new go
>> server.
>> >
>> > http://goratingserver.appspot.com
>>
>>   I tried a game last evening too.  I didn't mind not being able to
>> choose my nickname too much (I just clicked until I got a rodent nick ;),
>> and I liked the mechanics and minimalism in general.
>>
>>   However, at the end of the game both me and my opponent became a bit
>> confused as we didn't realize we have to capture dead stones; I'd
>> strongly suggest adding ability to mark dead stones before releasing it
>> to wider audience; playing it out is really tedious and my opponent
>> didn't realize they had to do it at all (the message is not so well
>> visible, they might not know English or even know how to read).
>>
>>   I'm not sure if I'll return; if I wanted to play in the browser, I'd
>> pick OGS and it doesn't seem clear what the advantages of your server
>> are, possibly besides your unique rating algorithm (which I'd also love
>> to hear more about).
>>
>> --
>> Petr Baudis
>> If you have good ideas, good data and fast computers,
>> you can do almost anything. -- Geoffrey Hinton
>> ___
>> Computer-go mailing list
>> Computer-go@computer-go.org
>> http://computer-go.org/mailman/listinfo/computer-go
>>
>
> ___
> Computer-go mailing list
> Computer-go@computer-go.org
> http://computer-go.org/mailman/listinfo/computer-go
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] new bot friendly go server.

[daniel rich]

Nice, Thanks Henry!

I am in the process of bringing up my first bot(as are a bunch of people)
so I will probably try and connect in the next couple of days.

--Danny

On Mon, May 23, 2016 at 8:46 AM, Petr Baudis  wrote:

>   Hi!
>
> On Sun, May 22, 2016 at 08:56:09AM +, Henry Hemming wrote:
> > Hello, I would like to invite all you go bot developers to my new go
> server.
> >
> > http://goratingserver.appspot.com
>
>   I tried a game last evening too.  I didn't mind not being able to
> choose my nickname too much (I just clicked until I got a rodent nick ;),
> and I liked the mechanics and minimalism in general.
>
>   However, at the end of the game both me and my opponent became a bit
> confused as we didn't realize we have to capture dead stones; I'd
> strongly suggest adding ability to mark dead stones before releasing it
> to wider audience; playing it out is really tedious and my opponent
> didn't realize they had to do it at all (the message is not so well
> visible, they might not know English or even know how to read).
>
>   I'm not sure if I'll return; if I wanted to play in the browser, I'd
> pick OGS and it doesn't seem clear what the advantages of your server
> are, possibly besides your unique rating algorithm (which I'd also love
> to hear more about).
>
> --
> Petr Baudis
> If you have good ideas, good data and fast computers,
> you can do almost anything. -- Geoffrey Hinton
> ___
> Computer-go mailing list
> Computer-go@computer-go.org
> http://computer-go.org/mailman/listinfo/computer-go
>
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] new bot friendly go server.

[Petr Baudis]

  Hi!

On Sun, May 22, 2016 at 08:56:09AM +, Henry Hemming wrote:
> Hello, I would like to invite all you go bot developers to my new go server.
> 
> http://goratingserver.appspot.com

  I tried a game last evening too.  I didn't mind not being able to
choose my nickname too much (I just clicked until I got a rodent nick ;),
and I liked the mechanics and minimalism in general.

  However, at the end of the game both me and my opponent became a bit
confused as we didn't realize we have to capture dead stones; I'd
strongly suggest adding ability to mark dead stones before releasing it
to wider audience; playing it out is really tedious and my opponent
didn't realize they had to do it at all (the message is not so well
visible, they might not know English or even know how to read).

  I'm not sure if I'll return; if I wanted to play in the browser, I'd
pick OGS and it doesn't seem clear what the advantages of your server
are, possibly besides your unique rating algorithm (which I'd also love
to hear more about).

-- 
Petr Baudis
If you have good ideas, good data and fast computers,
you can do almost anything. -- Geoffrey Hinton
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] new bot friendly go server.

[Rémi Coulom]

Hi Henry,

Thanks for your message. I tried your server (as a human player, not a bot). It 
might become an interesting alternative to CGOS, mixing humans and bots 
together.

I find it annoying that I cannot choose my pseudo. I understand that you might 
not want to worry about moderating offensive pseudos. But it is really 
frustrating.

Also, I would like to be able to watch games played by others.

I (GentleWindCrow) just played a game against "CurlyLion". After a while, it 
stopped playing. I believe I might have lost on time. But there was no message, 
and the game became blocked.

It would be nice to indicate whether the opponent is bot or human.

I am curious about your rating system, too.

Rémi

- Mail original -
De: "Henry Hemming" 
À: computer-go@computer-go.org
Envoyé: Dimanche 22 Mai 2016 10:56:09
Objet: [Computer-go] new bot friendly go server.



Hello, I would like to invite all you go bot developers to my new go server. 


http://goratingserver.appspot.com 


Here are a few reasons why its bot friendly. Its a simple (and very pretty) 
auto match server, with no handicaps, and only such pairings that there is a 
reasonable chance for both players to win. The only rule-set is Chinese and 
there is no arbitration in scoring, server provides estimate on which stones 
are dead, if either player thinks its wrong, they play cleanup, removing all 
dead stones from the board. Clock runs even during scoring, and there is no 
disconnect protection, ensuring all games finish in a timely manner. There are 
no restrictions against bots playing each other, however pairing algorithm 
tries to avoid repeat pairings. Accounts rating are much more accurate than 
standard elo yet will never get "heavy" allowing for quick testing of bot 
features. All games are ranked and there is no special registration 
requirements for bots. 


While there are hundreds of games being played every day by beta testers 
(ranging from double digit kyus to high dans), the server is still in stealth 
mode of sorts. I would like to give any bots a few days head start before 
opening it to the public. The server provides a RESTful interface, but there is 
a simple GTP translation code available for it, which is used to run Pachi and 
GnuGo at the moment. 


https://github.com/typohh/GTPRest 



-Henry Hemming 
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] Selfplay Phenomena

[Gian-Carlo Pascutto]

On 23-05-16 13:57, "Ingo Althöfer" wrote:
> Hi Gian-Carlo,
> 
>> Unsurprisingly, self-play favors extreme selectivity, but this does not
>> hold against other opponents.
> 
> is this just your personal experience, or are there systematic experiments on 
> this?
> Is it true "only" for MCTS (and vairants) or also for game tree search in 
> chess?

It's only personal experience. The only reason to have experience with
self-play behavior is a lack of suitable opponents to perform large
scale experiments with.

This was never the case in chess, so I never tried there.

-- 
GCP
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] Selfplay Phenomena

[Ingo Althöfer]

Hi Gian-Carlo,

> Unsurprisingly, self-play favors extreme selectivity, but this does not
> hold against other opponents.

is this just your personal experience, or are there systematic experiments on 
this?
Is it true "only" for MCTS (and vairants) or also for game tree search in chess?

Ingo.
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] Hajin Lee will play a live commented game against CrazyStone

[Gian-Carlo Pascutto]

On 22/05/2016 23:07, Álvaro Begué wrote:
> Disclaimer: I haven't actually implemented MCTS with NNs, but I have
> played around with both techniques.
> 
> Would it make sense to artificially scale down the values before the
> SoftMax is applied, so the probability distribution is not as
> concentrated, and unlikely moves are not penalized as much?

Not really, you can just treat the values as priors in UCT, and it's a
classical exploration/exploitation dilemma again. No need to fiddle with
the NN itself. You have the search exactly to find whether the NN is
wrong, after all.

Unsurprisingly, self-play favors extreme selectivity, but this does not
hold against other opponents. Wonder if anything went wrong there for CS :-)

-- 
GCP
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] Hajin Lee will play a live commented game against CrazyStone

[Rémi Coulom]

I think one of the main problems is that the network learns good replies 
to good moves. The training set does not have good replies to bad moves, 
but the search tree is full of bad moves that need to be punished.


Alvaro's suggestion looks good. This is one of the experiments I want to 
try.


Rémi

On 05/22/2016 11:41 PM, Henry Hemming wrote:
If the network is too selective, the cost function used to generate it 
doesn't penalize extreme predictions sufficiently? It was generated 
using quadratic cost when it should have been using cross-entropy cost?


On Mon, May 23, 2016 at 12:08 AM Álvaro Begué > wrote:


Disclaimer: I haven't actually implemented MCTS with NNs, but I
have played around with both techniques.

Would it make sense to artificially scale down the values before
the SoftMax is applied, so the probability distribution is not as
concentrated, and unlikely moves are not penalized as much?



On Sun, May 22, 2016 at 3:54 PM, Rémi Coulom > wrote:

Hi,

Thanks for using Crazy Stone.

I tried changes during the week, but nothing worked. So the
version that played the game was almost identical to the
commercial version.

The search did not anticipate Black E8 after B3. It seems the
NN makes the search too selective. I will investigate more.

Rémi

- Mail original -
De: "Paweł Morawiecki" >
À: computer-go@computer-go.org

Envoyé: Dimanche 22 Mai 2016 21:29:56
Objet: Re: [Computer-go] Hajin Lee will play a live commented
game against  CrazyStone



Hi,









It's fun to hear the pro making comments as she goes. I had
hoped for a better game, though.
Any comments from the CS camp?



I'm not from CrazyStone Team but a happy user of CS Deep Learning.


I analyzed the game (30 000 playouts per move) with the
version commercially available and it got everything right. I
mean every move Hajin Lee questioned was also questioned by
CrazyStone running on my PC. It includes:


- questionable attachment in the first joseki they played
- hane in upper-left corner
- and finally a blunder at b3 (loosing move) and selection of
this joseki as well


Remi said that he's been working hard to improve CS over the
last week, but it looks like something went wrong and instead
he got clearly a weaker program. Particularly this B3 is
really strange, where right after this white catches
everything with one move. Remi, what went wrong?


Cheers,
Paweł













Thanks,
Álvaro.








On Mon, May 16, 2016 at 3:58 AM, Xavier Combelle <
xavier.combe...@gmail.com  >
wrote:



That's fantastic


I suppose crazystone will play with crazystone account, but
what will be her handle ?





2016-05-16 9:50 GMT+02:00 Rémi Coulom < remi.cou...@free.fr
 > :


Hi,

I am very happy to announce that Hajin Lee will play a live
commented game against Crazy Stone on Sunday, at 8PM Korean
time. The game will take place on KGS, and she will make live
comments on her youtube channel.

Haylee's youtube:
https://www.youtube.com/c/HayleesWorldofGoBaduk

Rémi
___
Computer-go mailing list
Computer-go@computer-go.org 
http://computer-go.org/mailman/listinfo/computer-go

___
Computer-go mailing list
Computer-go@computer-go.org 
http://computer-go.org/mailman/listinfo/computer-go


___
Computer-go mailing list
Computer-go@computer-go.org 
http://computer-go.org/mailman/listinfo/computer-go


___
Computer-go mailing list
Computer-go@computer-go.org 
http://computer-go.org/mailman/listinfo/computer-go
___
Computer-go mailing list
Computer-go@computer-go.org 
http://computer-go.org/mailman/listinfo/computer-go


___
Computer-go mailing list
Computer-go@computer-go.org