Re: [computer-go] Hydra theory (was Hybrid theory)

http://www.robustdecisions.com/bayesianmethoddecisions.pdf This paper covers something called "robust decision theory" ... looks like it tackles similar problems to those under discussion. Terry McIntyre <[EMAIL PROTECTED]> ___

Re: [computer-go] Hydra theory (was Hybrid theory)

On 1, Feb 2008, at 7:31 PM, Don Dailey wrote: terry mcintyre wrote: ... From Don and Terry's comments and suggestions it should be obvious that the answers are not so obvious, and thus this is a reasonable subject for research. I appreciate these pointers to voting theory, and I am also rea

Re: [computer-go] Hydra theory (was Hybrid theory)

On 1, Feb 2008, at 5:39 PM, terry mcintyre wrote: "Hydra" ( apt choice, David! ) While I at first intended it as a minor pun on Hybrid, when reading your reply I realized that there may have been a subconscious nod to Chrilly Donninger and his Hydra chess program. So, I will acknowledge his

Re: [computer-go] Hydra theory (was Hybrid theory)

> Here's another approach: "Range Voting" > > http://rangevoting.org/rangeVborda.html > > The author of this particular page makes much of the pitfalls of strategic > voting, which should not matter to a set of emotionally disinterested, > independent agent routines. But range voting has one fur

Re: [computer-go] Hydra theory (was Hybrid theory)

>> The author of this particular page makes much of the pitfalls of strategic >> voting, which should not matter to a set of emotionally disinterested, >> independent agent routines. But range voting has one further advantage over >> borda voting: expressiveness. If an agent is given 99 votes t

Re: [computer-go] Hydra theory (was Hybrid theory)

terry mcintyre wrote: > > Here's another approach: "Range Voting" > > > http://rangevoting.org/rangeVborda.html > That particular article doesn't come across as being very scholarly. I would much prefer to see a good quality paper on it. Take this with a grain of salt, but here is my ow

Re: [computer-go] Hydra theory (was Hybrid theory)

My main point is that voting is a pandora's box. You make it sound like anything goes but most tries and intuitions will be broken. So I still suggest you start with something that is known to have good theoretical properties and use that as your base-line. Voting theory is not ad-hoc, it h

Re: [computer-go] Hydra theory (was Hybrid theory)

There is a probabilistic form of borda voting, which might be suitable for playouts, or deciding which nodes of the tree to expand first: http://www.springerlink.com/content/7xr1kleh0ktqg54y/ Never m

Re: [computer-go] Hydra theory (was Hybrid theory)

From: Don Dailey <[EMAIL PROTECTED]> > But take care - it's my understanding that borda relies totally on > ranking ALL choices and that to work correctly every agent much vote on > ALL candidates, ranking them from best to worst. > Voting theory is like hashing

Re: [computer-go] Hydra theory (was Hybrid theory)

Thanks to both Don and David for their comments. To add one idea: voting theory per se attempts to balance additional criteria, such as prevention of strategic voting, which don't concern us in designing a game-playing program, since the multiple agents of a Go "Hydra" ( apt choice, David! ) a

[computer-go] bigMogo_16

FYI :We just added a program called "bigMogo_16" to the scalability study.This is basically a control to see if mogo performs significantly better when more nodes are allocated per our earlier discussions. bigMogo_16 is Mogo_16 with 1,600,000 allocated instead of the default 400,000, an

[computer-go] Re: Move Prediction and Strength in Monte-Carlo Go

Hello. >Hi, >Sorry for announcing too early. Your English is maybe a bit exotic, but not too difficult to understand. I appreciate your effort to write in English. In the computer science course that I belong to, we have to write master thesis in English (even graduation thesis). >Also, I b

Re: [computer-go] Hydra theory (was Hybrid theory)

David Doshay wrote: > I looked up borda voting on Wikipedia. I did not know this was called > Borda voting, and it might be called a zeroth-order version of what I > am thinking. Rather than just take rank order from each, I intended to > try to include other metrics, for example, some measure of

[computer-go] Hydra theory (was Hybrid theory)

I looked up borda voting on Wikipedia. I did not know this was called Borda voting, and it might be called a zeroth-order version of what I am thinking. Rather than just take rank order from each, I intended to try to include other metrics, for example, some measure of distance from top. One engin

Re: [computer-go] Hybrid theory

Apologies for not quoting Don Dailey's text on Borda voting -- yahoo is doing something truly awful with quoted text, for some reason. David Doshay has mentioned both on-list and in offline conversation that one hurdle faced by multi-agent intelligences is the varying scales used to rank moves.

Re: [computer-go] Transformation between KGS ratings and Elo

On Sat, 2 Feb 2008, Alain Baeckeroot wrote: 1800 is gnugo, so this puts top programs near 1k (2d for extreme mogo_18) this seems reasonable to me. Are you confusing 19x19 and 9x9? The ELO/KGS table is for 19x19, while mogo_18 plays 9x9. Christoph ___

Re: [computer-go] scalability study

Le vendredi 1 février 2008, terry mcintyre a écrit : > Regarding the scalability study, ... > I'm very curious about that flat spot for > Mogo-16, 17, and 18. ( http://cgos.boardspace.net/study/index.html ) > I think its just lack of data Mogo_16 = 2958+47 / -45 Mo

Re: [computer-go] Transformation between KGS ratings and Elo

Le vendredi 1 février 2008, Andy a écrit : > See below I created a table that shows the transformation from KGS ratings > to the Elo that CGOS uses. I set 6k=1800 because I believe that is what GNU > 3.7.10 is under both systems. Does anyone have more data points for bots > that play on both syst

Re: [computer-go] 19x19 Study

Jason House wrote: > > > On Jan 29, 2008 10:16 AM, Jason House <[EMAIL PROTECTED] > > wrote: > > > > On Jan 29, 2008 10:01 AM, Don Dailey <[EMAIL PROTECTED] > > wrote: > > FatMan seems to hit some kind of hard limit rather sudden

Re: [computer-go] Hybrid theory

terry mcintyre wrote: > UCT is based on a theory of a multi-armed bandit, with uncertain knowledge > about which "arms" would be most productive. Is it possible to graft various > sources of knowledge into a sort of meta-bandit algorithm? > > As to fusing top-level knowledge with random playout

Re: [computer-go] Transformation between KGS ratings and Elo

Andy wrote: > See below I created a table that shows the transformation from KGS > ratings to the Elo that CGOS uses. I set 6k=1800 because I believe > that is what GNU 3.7.10 is under both systems. Does anyone have more > data points for bots that play on both systems? > > Also is there an "al

Re: [computer-go] Hybrid theory

I'm not expert on decision theory, but it's my understanding that borda counting or voting is excellent way to integrate different decision making agents.Of course this depends a lot on the nature of the decision to be made, but if you have N choices and several agents that are capable of rank

Re: [computer-go] Hybrid theory

Yes, I would like to see the references. I also wonder how similar the statistical estimators are to each other. Do they represent data and estimates in the same way? We are trying to move towards a design that mixes intrinsically different kinds of engines. Neural Networks, UCT/MC, Gnu Go, tile/

Re: [computer-go] 19x19 Study

On Jan 29, 2008 10:16 AM, Jason House <[EMAIL PROTECTED]> wrote: > > > On Jan 29, 2008 10:01 AM, Don Dailey <[EMAIL PROTECTED]> wrote: > > > FatMan seems to hit some kind of hard limit rather suddenly. It > > could be an implementation bug or something else - I don't really > > understand this

Re: [computer-go] Hybrid theory

On Feb 1, 2008 4:18 PM, terry mcintyre <[EMAIL PROTECTED]> wrote: > UCT is based on a theory of a multi-armed bandit, with uncertain knowledge > about which "arms" would be most productive. Is it possible to graft various > sources of knowledge into a sort of meta-bandit algorithm? IIRC, MoGo al

Re: [computer-go] Hybrid theory

UCT is based on a theory of a multi-armed bandit, with uncertain knowledge about which "arms" would be most productive. Is it possible to graft various sources of knowledge into a sort of meta-bandit algorithm? As to fusing top-level knowledge with random playouts, I love the idea, and am tryin

[computer-go] Transformation between KGS ratings and Elo

See below I created a table that shows the transformation from KGS ratings to the Elo that CGOS uses. I set 6k=1800 because I believe that is what GNU 3.7.10 is under both systems. Does anyone have more data points for bots that play on both systems? Also is there an "all times" list for 19x19?

Re: [computer-go] Hybrid theory

Le vendredi 1 février 2008, David Doshay a écrit : > This is the direction in which we are moving with SlugGo. We also > expect it to be difficult to integrate different approaches, but this > has always been our research direction: when there are multiple > codes which will each give an evaluation

Re: [computer-go] Hybrid theory

Michael Williams wrote: So do I. I just stated a simpler version here because I previously suggested a more integrated approach and got zero replies. I'll state it again: Start with a UCT+MC engine. When a tree node reaches X number of playouts (1000?, 1?), do a tactical analysis. Wh

Re: [computer-go] Hybrid theory

So do I. I just stated a simpler version here because I previously suggested a more integrated approach and got zero replies. I'll state it again: Start with a UCT+MC engine. When a tree node reaches X number of playouts (1000?, 1?), do a tactical analysis. When that analysis exists for

Re: [computer-go] Hybrid theory

This is the direction in which we are moving with SlugGo. We also expect it to be difficult to integrate different approaches, but this has always been our research direction: when there are multiple codes which will each give an evaluation of a situation, how does one design an arbitrator that ma

Re: [computer-go] Hybrid theory

There is much to think about with Jason's and Michaels ideas.I favor a more integrated approach than Michael suggests because I think it would be very difficult to essentially have 2 different programs playing the same game (ever play non-consultation doubles in chess or go? It's fun but the

Re: [computer-go] Hybrid theory

I wouldn't stop there. I'd like a static analyzer to add tactical smarts to playouts. If there's a pre-existing nakade, seki, etc, the playouts should get it right. On Feb 1, 2008 10:34 AM, Michael Williams <[EMAIL PROTECTED]> wrote: > I think we would all agree that UCT+MC is quite good for st

[computer-go] Hybrid theory

I think we would all agree that UCT+MC is quite good for strategy but not so good with tactics. I'd like to see this hybrid engine: One that starts with a traditional full-board static analysis (with local tactical searches), looking for urgent moves. If it finds an urgent move, it makes it. I

Re: [computer-go] UCT and solving life and death

David Fotland wrote: Since you hijacked my thread, I'm changing the title and injecting some data :) I tried to be very clear that I didn't want that thread to become another scalability flame fest. Here is a high level mogo game, level 15 vs level 16, that hinges on a life and death problem th

[computer-go] CGOS 19x19

Cgos 19x19 needs some work due to the change in our website. It will probably not work before a few days (monday or tuesday it should be ok). Olivier Same error here. Terry McIntyre <[EMAIL PROTECTED]> ÿÿWherever is found what is called a paternal government, there

Re: [computer-go] CGos 19 cannot view game records

Same error here. Terry McIntyre <[EMAIL PROTECTED]> “Wherever is found what is called a paternal government, there is found state education. It has been discovered that the best way to insure implicit obedience is to commence tyranny in the nursery.” Benjamin Disraeli, Speech in the House of

[computer-go] CGos 19 cannot view game records

Is just my problem or do others have it. When I try to download for example: http://www.lri.fr/~teytaud/SGF/2008/01/30/17067.sgf I Get 403 response with following line: You don't have permission to access /~teytaud/SGF/2008/01/30/17067.sgf on this server. Cheers, Petri -- Petri Pitkänen e-mail: [

Re: [computer-go] Re: Move Prediction and Strength in Monte-Carlo Go Program

荒木伸夫 wrote: Hello, Coulom. I'm Nobuo Araki. Thank you for reading my thesis. However, this thesis is first version, not final version. Therefore, there are too few experiments. And Mr. Hideki Kato sent me many warnings about this thesis, for example "English is too bad." You may be confused w