[Computer-go] English version WGC live is here
https://www.youtube.com/watch?v=xnm_o-ymUgU=youtu.be Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Crazy Stone is back
valky...@phmp.se: <19f31e7e5cdf310b9afa91f577997...@phmp.se>: >I think you misunderstood what I wrote, >if perfect play on 9x9 is 6000 Elo, then if the value function is 3000 >Elo and MC eval is 2000 Elo with 1 second thinking time then it might >be that the combination of a value function and MC eval ends up being >2700 Elo. It could also be that it ends up at 3200 Elo. > >I personally believe MC eval alone can be very strong so it might be >that the capacity of a neural network is not enough to replace MC eval. > >When I wrote "converge to perfection" it was not to claim that the Zero >approach reaches perfection, just that it get stronger over time. How to gurantee the improvement does not stop nor oscillate? Actually, the first instance (40 layer version) of AlphaGo Zero stopped improvements in three days (at least looks so). >The interesting question is if old school MC evaluation can fill up the >knowledge gaps of the value function. My point is that. As value networks approximate the value function in vary rough manner (ie, smoother parts) due to not enough freedom and/or samples and MC rollouts can implement (maybe partly) the detail parts of the function, mixing those two could yields better approximation (in theory :). Hideki >For my own Odin project I am not working on the MC evaluation currently, >since only when I a have final neural network solution can I see which >weaknesses I need to fix. > >Best >Magnus > > >On 2018-03-05 21:12, Hideki Kato wrote: >> DCNNs are not magic but just non-linear continuous function >> approximators with finite freedom and we can provide up to >> 10^8 samples (board positions) in practice. >> >> Why do most people believe VN can approximate (perfect or >> near perfect) value function? What do they estimate the >> complexity of the value function for 19x19 Go? >> >> valky...@phmp.se: <aa3cca138c40cbd620700cc36e950...@phmp.se>: >>> My guess is that there is some kind of threshold depending on the >>> relative strength of MC eval and the value function of the NN. >>> >>> If the value function is stronger than MC eval I would guess MCEval >>> turns into a bad noisy feature with little benefit. >>> >>> Depending on how strong MC eval is this threshold is probably very >>> different between engines. Also i can imagine that NN value function >>> can >>> have some gaping holes in its knowledge that even simple MC eval can >>> patch up. Probably true for supervised learning where training data >>> probably has a lot of holes since bad moves are not in the data. >>> >>> The Zero approach is different because it should converge to >>> perfection >>> in the limit, thus overcome any weaknesses of the value function early >>> on. At least in theory. >>> >>> >>> On 2018-03-05 14:04, Gian-Carlo Pascutto wrote: >>>> On 5/03/2018 12:28, valky...@phmp.se wrote: >>>>> Remi twittered more details here (see the discussion with gghideki: >>>>> >>>>> https://twitter.com/Remi_Coulom/status/969936332205318144 >>>> >>>> Thank you. So Remi gave up on rollouts as well. Interesting >>>> "difference >>>> of opinion" there with Zen. >>>> >>>> Last time I tested this in regular Leela, playouts were beneficial, >>>> but >>>> this was before combined value+policy nets and much more training >>>> data >>>> was available. I do not know what the current status would be. >>> >>> ___ >>> 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Crazy Stone is back
DCNNs are not magic but just non-linear continuous function approximators with finite freedom and we can provide up to 10^8 samples (board positions) in practice. Why do most people believe VN can approximate (perfect or near perfect) value function? What do they estimate the complexity of the value function for 19x19 Go? valky...@phmp.se: <aa3cca138c40cbd620700cc36e950...@phmp.se>: >My guess is that there is some kind of threshold depending on the >relative strength of MC eval and the value function of the NN. > >If the value function is stronger than MC eval I would guess MCEval >turns into a bad noisy feature with little benefit. > >Depending on how strong MC eval is this threshold is probably very >different between engines. Also i can imagine that NN value function can >have some gaping holes in its knowledge that even simple MC eval can >patch up. Probably true for supervised learning where training data >probably has a lot of holes since bad moves are not in the data. > >The Zero approach is different because it should converge to perfection >in the limit, thus overcome any weaknesses of the value function early >on. At least in theory. > > >On 2018-03-05 14:04, Gian-Carlo Pascutto wrote: >> On 5/03/2018 12:28, valky...@phmp.se wrote: >>> Remi twittered more details here (see the discussion with gghideki: >>> >>> https://twitter.com/Remi_Coulom/status/969936332205318144 >> >> Thank you. So Remi gave up on rollouts as well. Interesting "difference >> of opinion" there with Zen. >> >> Last time I tested this in regular Leela, playouts were beneficial, but >> this was before combined value+policy nets and much more training data >> was available. I do not know what the current status would be. > >_______ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] 9x9 is last frontier?
Do you think deep learning can understand and solve double ko, for example? Hideki Aja Huang:
Re: [Computer-go] 9x9 is last frontier?
uurtamo .: <cadg0inbjvu2qzahgkyym+ahobqdv5rmwgcnqttcxqkzhqfs...@mail.gmail.com>: >I didn't mean to suggest that I can or will solve this problem tomorrow. > >What I meant to say is that it is clearly obvious that 9x9 is not immune to >being destroyed -- it's not what people play professionally (or at least is >not what is most famous for being played professionally), so it is going to >stand alone for a little while; it hasn't been the main focus yet. I >understand that it technically has features such as: very tiny point >differences; mostly being tactical. I don't think or have reason to believe >that that makes it somehow immune. > >What concerns me is pseudo-technical explanations for why it's harder to >beat humans at 9x9 than at 19x19. Saying that it's harder at 9x9 seems like >an excuse to explain (or hopefully justify) how the game is still in the >hands of humans. This feels very strongly like a justification for how "go >is still really hard for computers". Which, I suppose, we can break down >into lots of little subcases and worry about. The tiny point difference >issue is interesting; it means that things need to be super tight (less >room for sloppy play). Checkers also has this feature. > >The reality, in my unjustified opinion, is that this will be a solved >problem once it has obtained enough focus. I'm suspecious. The value network (VN) is not enough for 9x9 because VN can't approximate value functions at enough detail. This is also a problem on 19x19 but the advantages VN gives at silent positions is big enough (actually a few points) to beat top level human players. I believe another idea is necessary for 9x9. #One possible (?) simple solution: if the inference speed of the policy network gets 100 or more times faster then we can use PN directly in rollouts. This may make VN useless. Go is still hard for both human and computers :). Hideki >s. > > >On Fri, Feb 23, 2018 at 6:12 PM, Hideki Kato <hideki_ka...@ybb.ne.jp> wrote: > >> uurtamo .: > 1vhk7t...@mail.gmail.com>: >> >Slow down there, hombre. >> > >> >There's no secret sauce to 9x9 other than that it isn't the current focus >> >of people. >> > >> >Just like 7x7 isn't immune. >> > >> >A computer program for 9x9, funded, backed by halfway serious people, and >> >focused on the task, will *destroy* human opponents at any time it needs >> to. >> >> Why do you think (or believe) so? I'd like to say there >> is no evidence so far. >> >> >If you believe that there is a special reason that 9x9 is harder than >> >19x19, then I'm super interested to hear that. But it's not harder for >> >computers. It's just not what people have been focusing on. >> >> 9x9 is not harder than 19x19 as a game. However: (1) Value >> networks, the key components to beat human on 19x19, work >> fine only on static positions but 9x9 has almost no such >> positions. (2) Humans can play much better on 9x9 >> than 19x19. Top level professionals can read-out at near >> end of the middle stage of a game in less than 30 min with >> one point accuracy of the score, for example. >> >> Humans are not good at global evaluation of larger boards so >> bots can beat top professionals on 19x19 but this does not >> apply 9x9. The size of the board is important because >> value networks are not universal, ie, approximate the >> value function not so presicely, mainly due to >> the number of training data is limited in practice (up to >> 10^8 while the number of possible input positions is greater >> than, at least, 10^20). One more reason, there are no >> algorithm to solve double ko. This is not so big problem on >> 19x19 but 9x9. >> >> Best, Hideki >> >> >s. >> > >> >On Feb 23, 2018 4:49 PM, "Hideki Kato" <hideki_ka...@ybb.ne.jp> wrote: >> > >> >> That's not the point, Petri. 9x9 has almost no "silent" >> >> or "static" positons which value networks superb humans. >> >> On 9x9 boards, Kos, especially double Kos and two step Kos >> >> are important but MCTS still works worse for them, for >> >> examples. Human professionals are much better at life >> >> and complex local fights which dominate small board games >> >> because they can read deterministically and deeper than >> >> current MCTS bots in standard time settings (not blitz). >> >> Also it's well known that MCTS is not good at finding narrow >> >> and deep paths to win due to "averaging". Ohashi 6p said >> >&
Re: [Computer-go] Crazy Stone is back
A guidline for CGOS users: Do self-play to pick-up few strongest programs at your site (never use CGOS for this purpose) and throw them into CGOS to evaluate their ratings among others. Please note that CGOS is the valueable shared resource for all developers. #We (mainly Hiroshi and me) are managing CGOS in such a way that a new program can have solid rating as quick as possible (ie, fewer games). The best (quickest) case could be: one new (unrated) program and many anchor-like (ie, having stable rating) programs at every 200 to 300 Elo. In this case, every game gives much information to estimate the rating of the new program. More new, unrated programs slower the convergence. Flooding is a very bad idea for all. Best, Hideki Hiroshi Yamashita: <35aed3ab-235f-adc2-b072-1b317982a...@bd.mbn.or.jp>: >No Zen on CGOS is pity. > >To LZ-0xx-p1600-t1-r1 author, >I think LZ-073-p1600-t1-r1 has BayesElo already. > > From LeelaZero page, >73 2018-02-05 23:0654bfb7b8 >LZ-54bfb7-t1-p1600, BayesElo is 2903. > >Recalculating CGOS rating is not essential. >And too many same kind bots running makes many selfplay matching. >Its rating is more unreliable. >Could you stop them, and run up to two or three bots which has no BayesElo? > >Thanks, >Hiroshi Yamashita > > >On 2018/02/28 17:12, Hideki Kato wrote: >> Welcome back Remi! >> >> On the 19x19 cgos, recently many LeelaZeros are running. >> This flood is making CGOS less useful and so I'll reconnect >> Zen after the flooding ends. Sorry for inconvinience. >> >> Hideki >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Crazy Stone is back
Welcome back Remi! On the 19x19 cgos, recently many LeelaZeros are running. This flood is making CGOS less useful and so I'll reconnect Zen after the flooding ends. Sorry for inconvinience. Hideki Remi Coulom: <1656424330.7641525.1519798424452.javamail.r...@spooler6-g27.priv.proxad.net>: >Hi, > >I have just connected the newest version of Crazy Stone to CGOS. It is >based on the AlphaZero approach. The "Weights" engine were in fact previous >experimental versions. CrazyStone-18.03 is using time control and pondering >instead of a fixed number of evaluations per move. So it should be much >stronger than Weights_31_3200. > >Does anybody know who cronus is? It is _extremely_ strong. Its rating is >low because it has had only weaker opponents, but it is undefeated so far, >except for one loss on time, and some losses against other versions of >itself. It has just won two games in a row against Crazy Stone. > >I hope the other strong engines will reconnect, too. > >Rémi >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] 9x9 is last frontier?
uurtamo .: <cadg0incam-_ih31cbka4mvg2fbdmj3adqwcuyfxtb1vhk7t...@mail.gmail.com>: >Slow down there, hombre. > >There's no secret sauce to 9x9 other than that it isn't the current focus >of people. > >Just like 7x7 isn't immune. > >A computer program for 9x9, funded, backed by halfway serious people, and >focused on the task, will *destroy* human opponents at any time it needs to. Why do you think (or believe) so? I'd like to say there is no evidence so far. >If you believe that there is a special reason that 9x9 is harder than >19x19, then I'm super interested to hear that. But it's not harder for >computers. It's just not what people have been focusing on. 9x9 is not harder than 19x19 as a game. However: (1) Value networks, the key components to beat human on 19x19, work fine only on static positions but 9x9 has almost no such positions. (2) Humans can play much better on 9x9 than 19x19. Top level professionals can read-out at near end of the middle stage of a game in less than 30 min with one point accuracy of the score, for example. Humans are not good at global evaluation of larger boards so bots can beat top professionals on 19x19 but this does not apply 9x9. The size of the board is important because value networks are not universal, ie, approximate the value function not so presicely, mainly due to the number of training data is limited in practice (up to 10^8 while the number of possible input positions is greater than, at least, 10^20). One more reason, there are no algorithm to solve double ko. This is not so big problem on 19x19 but 9x9. Best, Hideki >s. > >On Feb 23, 2018 4:49 PM, "Hideki Kato" <hideki_ka...@ybb.ne.jp> wrote: > >> That's not the point, Petri. 9x9 has almost no "silent" >> or "static" positons which value networks superb humans. >> On 9x9 boards, Kos, especially double Kos and two step Kos >> are important but MCTS still works worse for them, for >> examples. Human professionals are much better at life >> and complex local fights which dominate small board games >> because they can read deterministically and deeper than >> current MCTS bots in standard time settings (not blitz). >> Also it's well known that MCTS is not good at finding narrow >> and deep paths to win due to "averaging". Ohashi 6p said >> that he couldn't lose against statiscal algorithms after the >> event in 2012. >> >> Best, >> Hideki >> >> Petri Pitkanen: <CAMp4Doefkp+n16CxDWY9at9OFwdh3V7+ >> 3zrby3k9kjvmzah...@mail.gmail.com>: >> >elo-range in 9x9 smaller than 19x19. One just cannot be hugelyl better >> than >> >the other is such limitted game >> > >> >2018-02-23 21:15 GMT+02:00 Hiroshi Yamashita <y...@bd.mbn.or.jp>: >> > >> >> Hi, >> >> >> >> Top 19x19 program reaches 4200 BayesElo on CGOS. But 3100 in 9x9. >> >> Maybe it is because people don't have much interest in 9x9. >> >> But it seems value network does not work well in 9x9. >> >> Weights_33_400 is maybe made by selfplay network. But it is 2946 in >9x9. >> >> Weights_31_3200 is 4069 in 19x19 though. >> >> >> >> In year 2012, Zen played 6 games against 3 Japanese Pros, and lost by >> 0-6. >> >> And it seems Zen's 9x9 strength does not change big even now. >> >> http://computer-go.org/pipermail/computer-go/2012-November/005556.html >> >> >> >> I feel there is still enough chance that human can beat best program in >> >> 9x9. >> >> >> >> Thanks, >> >> Hiroshi Yamashita >> >> >> >> ___ >> >> Computer-go mailing list >> >> Computer-go@computer-go.org >> >> http://computer-go.org/mailman/listinfo/computer-go >> > inline file >> >___ >> >Computer-go mailing list >> >Computer-go@computer-go.org >> >http://computer-go.org/mailman/listinfo/computer-go >> -- >> Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> >> ___ >> Computer-go mailing list >> Computer-go@computer-go.org >> http://computer-go.org/mailman/listinfo/computer-go > inline file >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] 9x9 is last frontier?
That's not the point, Petri. 9x9 has almost no "silent" or "static" positons which value networks superb humans. On 9x9 boards, Kos, especially double Kos and two step Kos are important but MCTS still works worse for them, for examples. Human professionals are much better at life and complex local fights which dominate small board games because they can read deterministically and deeper than current MCTS bots in standard time settings (not blitz). Also it's well known that MCTS is not good at finding narrow and deep paths to win due to "averaging". Ohashi 6p said that he couldn't lose against statiscal algorithms after the event in 2012. Best, Hideki Petri Pitkanen: <camp4doefkp+n16cxdwy9at9ofwdh3v7+3zrby3k9kjvmzah...@mail.gmail.com>: >elo-range in 9x9 smaller than 19x19. One just cannot be hugelyl better than >the other is such limitted game > >2018-02-23 21:15 GMT+02:00 Hiroshi Yamashita <y...@bd.mbn.or.jp>: > >> Hi, >> >> Top 19x19 program reaches 4200 BayesElo on CGOS. But 3100 in 9x9. >> Maybe it is because people don't have much interest in 9x9. >> But it seems value network does not work well in 9x9. >> Weights_33_400 is maybe made by selfplay network. But it is 2946 in 9x9. >> Weights_31_3200 is 4069 in 19x19 though. >> >> In year 2012, Zen played 6 games against 3 Japanese Pros, and lost by 0-6. >> And it seems Zen's 9x9 strength does not change big even now. >> http://computer-go.org/pipermail/computer-go/2012-November/005556.html >> >> I feel there is still enough chance that human can beat best program in >> 9x9. >> >> Thanks, >> Hiroshi Yamashita >> >> ___ >> Computer-go mailing list >> Computer-go@computer-go.org >> http://computer-go.org/mailman/listinfo/computer-go > inline file >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
[Computer-go] CLeelaz9Test1 on CGOS 9x9
Hi, CLeelaz9Test1 plays pass without clean-up dead stones. This makes server mis-score and causes wrong result. Please capture all dead stones before play pass. Ref. "CGOS Rules of Play" section at http://www.yss-aya.com/cgos/ . Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Zen giving handicap against pro players
Ingo Althofer: <trinity-4182fba1-fea1-435b-b80b-446da10aaa1e-1516710917576@3c-app-gmx-bs15>: >Dear Hideki, > >>>> On Yugen-no-ma, Nihonkiin's online Go site, many (mainly >>>> young) professionals have played more than 200 two-hc games >>>> against Zen and its winrate is greater than 70%, AFICR. >> >>>* Is that the "normal" Zen version, or one that is >>>specially tuned for handicap games? >> >> Not special. > >So, were the handicap games played at "normal" komi values >(6,5 or 7,5 points) ? No, but "standard" 0.5 pts. Zen supports very wide range of komi; about -20 to +30 (mainly for the users of Tencho-no-Igo). Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Zen giving handicap against pro players
>> BTW, the games between Zen and professionals can be watched >> any payed members of Nihonkiin. > >Thanks for that information. >Does the server has the opportunity for English interface? My friend told me that perhaps WBaduk client provides English interface for Yugen-no-Ma (except from Japan, Korea or China). http://www.wbaduk.com/ Register (free), download installer, and install the client "WBaduk." Then run WBaduk (on desktop by default) and choose Japanese server, although I didn't confirm. Hope this helps, Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Zen giving handicap against pro players
Dear Ingo, Ingo Althofer: <trinity-807596eb-8cb2-4db4-a450-9ac02cc7a1be-1516597193617@3c-app-gmx-bs66>: >Dear Hideki, > >> >> On Yugen-no-ma, Nihonkiin's online Go site, many (mainly >> >> young) professionals have played more than 200 two-hc games >> >> against Zen and its winrate is greater than 70%, AFICR. >> >> ... >> >> Early December or late November, I remember but due to an >> accident. HC games were prohibited but a young >> professional could play due to a bug (:-) of the server. >> After that, the manager changed the mind. > >Sometimes, little errors in reality forces decisions ;-) > >[We had it in Germany back on November 9, 1989, when Guenther >Schabowski was asked when the new travel reguations >for Eastern German citizens would become effective. He did >not know, stuttered around for some moment and then said: >"as far as I know, it will be effective from today on" - >and the wall was down.) Wow, what a historically big error! >> A professional (director of Japanese national team) told me >> that two hc games are more interesting and exciting to watch >> because Zen plays seriously :). > >Wow. > >> BTW, the games between Zen and professionals can be watched >> any payed members of Nihonkiin. > >Thanks for that information. >Does the server has the opportunity for English >interface? Hmmm, not now but I'll ask the manager later. Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Breakthrough: FineArt giving handicap against pro players
Dear Ingo, Ingo Althofer: <trinity-e14c760c-96ce-457d-bbee-b42e6ccf3312-1516568318166@3c-app-gmx-bs35>: >Hi Hideki, > >thanks for your posting. > >> On Yugen-no-ma, Nihonkiin's online Go site, many (mainly >> young) professionals have played more than 200 two-hc games >> against Zen and its winrate is greater than 70%, AFICR. > >Very interesting information, and congratulations >to Zen's performance. Thanks. >* Do you remember when Zen started to give handicaps >against pro players? Early December or late November, I remember but due to an accident. HC games were prohibited but a young professional could play due to a bug (:-) of the server. After that, the manager changed the mind. >* Is that the "normal" Zen version, or one that is >specially tuned for handicap games? Not special. >> #On even games (more than 2000), its recent winrate is about >> 98%. > >Having this in mind, it makes indeed sense that >Zen gives handicap stones. A professional (director of Japanese national team) told me that two hc games are more interesting and exciting to watch because Zen plays seriously :). BTW, the games between Zen and professionals can be watched any payed members of Nihonkiin. Hideki >Ingo. > >PS. When will AlphaGo start giving handicaps against pro players? In this year, I hope. >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Breakthrough: FineArt giving handicap against pro players
On Yugen-no-ma, Nihonkiin's online Go site, many (mainly young) professionals have played more than 200 two-hc games against Zen and its winrate is greater than 70%, AFICR. #On even games (more than 2000), its recent winrate is about 98%. Hideki Ingo Althofer: <trinity-475deaf6-c186-45dc-a589-2142fb59e928-1516554279098@3c-app-gmx-bs32>: >That is the new world: >Top computer programs are giving handicap to >(top) professional players! > >"We" had the intention to have a handicap setting >with DeepZen against a Japanese Pro in the European >Go Congress in July/August 2018, but the Japanese >organisation of Professional Go Players did not allow it. > >Now, the Chinese Go scene simply has created facts. > >Ingo. > >Von: "Jim O'Flaherty" <jim.oflaherty...@gmail.com> >It's unclear to me who played black with the two handicap stones. >Ke Jie or FineArt? >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Breakthrough: FineArt beating Ke Jie with 2 Handicap Stones
You have to edit every game record as follow: (1) add CA[utf-8] and (2) divide the value of KM[] by 100, ie, change from KM[650] to KM[6.5]. Hideki Ingo Althofer: <trinity-b395fe14-84ed-439e-bad8-51bcc5c4657f-1516520823075@3c-app-gmx-bs75>: >Stefan Kaitschick posted this in the German computer go forun: >http://www.dgob.de/yabbse/index.php?topic=6728.msg215694#msg215694 > >His text roughly translated to English: >> I downloaded FineArt free 2 stone games. >> At least the game of Ke Jies is included. >> Perhaps this is (essentially) the collection. >> You have to rename the downloaded file to new ending .zip > >Hope this helps, Ingo. > > >> Gesendet: Sonntag, 21. Januar 2018 um 05:49 Uhr >> Von: "Michael Alford" <m...@aracnet.com> >> An: computer-go@computer-go.org >> Betreff: Re: [Computer-go] Breakthrough: FineArt beating Ke Jie with 2 >Handicap Stones >> >> Could someone make sgf's of these games available? >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] AI ryusei 2017 first day result
I believe all game records will be uploaded to the offcial result page after cleanup (maybe in a few days), like UEC Cup. https://www.igoshogi.net/ai_ryusei/01/result.html Hideki Ray Tayek: <9c4b556b-45de-7f68-ee35-ce21749d6...@ca.rr.com>: >On 12/9/2017 1:36 AM, Hiroshi Yamashita wrote: >> Hi, >> >> 18 programs played swiss 7R, ... >> >> 15 programs will play tommorow tornament. >> >> Thanks, >> Hiroshi Yamashita >> > >does anyone know if we can get some of the .sgf files? > >thanks > >-- >Honesty is a very expensive gift. So, don't expect it from cheap people >- Warren Buffett >http://tayek.com/ >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] AI ryusei 2017 first day result
Official result <https://www.igoshogi.net/ai_ryusei/01/result.html> (in Japanese). Hideki Hiroshi Yamashita: <9D4FFB9B95534D1EBA18DE6C86502E57@x60>: >Hi, > >18 programs played swiss 7R, > > 1. DeepZenGo 6-1 > 2. FineArt6-1 > 3. DolBaram 5-2 > 4. Tianrang 5-2 > 5. Aya5-2 > 6. AQ 5-2 > 7. Maru 4-3 > 8. Abacus 4-3 > 9. Deep_ark 4-3 >10. SR Go 3-4 Category B >11. Raynz 3-4 >12. nlp3-4 >13. GNU Go 3-4 Guest >14. Kugutsu2-5 >15. Katsunari 2-5 >16. Kifuwarabe 2-5 Category B >17. KinoaIgo 1-6 >18. MayouiGo 0-7 > >15 programs will play tommorow tornament. > >Thanks, >Hiroshi Yamashita > >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Is MCTS needed?
>On 17-11-17 02:15, Hideki Kato wrote: >> Stephan K: >
Re: [Computer-go] Adding roll-out to Zero
Hi, Thank you for the info and an interesting idea. I wonder, however, if DCNN can replace handcraft rollouts... In the case of Zen, there are so many lines for special cases such as snapbacks, approach moves, and nakades to play "correct" reply moves at 100% probability. ReLU is not so good at approximating binary functions and, as a result, "dead" stones, for example, may live at a few tens percent in the average outcome of rollouts. Hideki patrick.bardou via Computer-go: <c3sejrforgume3r3vfhiuhpi.1510872947...@email.android.com>: >Hi Hideki, >I think they could have used a rollout policy network (RPN), as described >in "Convolutional Monte Carlo Rollouts in Go" >:https://arxiv.org/abs/1512.03375 >and have it trained based on the MCTS outcome, at the same time and in the >same way as the policy head is trained. This RPN would start playing random >rollout, then benefit from the policy head training. >This would let as "human knowledge" the mixing factor between rollout and >value net evaluations. But there is anyway such a mixing factor in Zero >training pipeline, in the loss function mixing policy and value heads. >Regards,Patrick > > > >Message: 1 >Date: Fri, 17 Nov 2017 02:32:29 +0900 >From: Hideki Kato <hideki_ka...@ybb.ne.jp> >To: computer-go@computer-go.org >Subject: Re: [Computer-go] Is MCTS needed? >Message-ID: <5a0dcba9.8060%hideki_ka...@ybb.ne.jp> >Content-Type: text/plain; charset=US-ASCII > >Hi, > >I strongly believe adding rollout makes Zero stronger. >They removed rollout just to say "no human knowledge". >#Though the number of past moves (16) has been tuned by >human :). > > > inline file >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Is MCTS needed?
Stephan K:
Re: [Computer-go] Is MCTS needed?
Hi, I strongly believe adding rollout makes Zero stronger. They removed rollout just to say "no human knowledge". #Though the number of past moves (16) has been tuned by human :). Hideki Petr Baudis: <20171116154309.tfq5ix2hzwzci...@machine.or.cz>: > Hi, > > when explaining AlphaGo Zero to a machine learning audience yesterday > > >(https://docs.google.com/presentation/d/1VIueYgFciGr9pxiGmoQyUQ088Ca4ouvEFDPoWpRO4oQ/view) > >it occurred to me that using MCTS in this setup is actually such >a kludge! > > Originally, we used MCTS because with the repeated simulations, >we would be improving the accuracy of the arm reward estimates. MCTS >policies assume stationary distributions, which is violated every time >we expand the tree, but it's an okay tradeoff if all you feed into the >tree are rewards in the form of just Bernoulli trials. Moreover, you >could argue evaluations are somewhat monotonic with increasing node >depths as you are basically just fixing a growing prefix of the MC >simulation. > > But now, we expand the nodes literally all the time, breaking the >stationarity possibly in drastic ways. There are no reevaluations that >would improve your estimate. The input isn't binary but an estimate in >a continuous space. Suddenly the Multi-armed Bandit analogy loses a lot >of ground. > > Therefore, can't we take the next step, and do away with MCTS? Is >there a theoretical viewpoint from which it still makes sense as the best >policy improvement operator? > > What would you say is the current state-of-art game tree search for >chess? That's a very unfamiliar world for me, to be honest all I really >know is MCTS... > >-- > Petr Baudis, Rossum > Run before you walk! Fly before you crawl! Keep moving forward! > If we fail, I'd rather fail really hugely. -- Moist von Lipwig >_______ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] November KGS bot tournament
Darren Cook: <92a6a6a9-4bb9-d8a8-fdca-4276a5ce1...@dcook.org>: >> Since AlphaGo, almost all academic organizations have >> stopped development but, ... > >In Japan, or globally? Either way, what domain(s)/problem(s) have they >switched into studying? Global. Some told me that there will be no value in developing strong programs, in academic sense. Maybe others have other reasons. The only exception I know is Prof. Wu's lab (CGI) in Taiwan. Some have shifted to generating natural language-like explanations for AI's moves, developing entertainment versions which intentionally play close games, etc. and some have shifted to other games including strategic ones and imperfect info ones. Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Zero is weaker than Master!?
Please read _through_ the paper sequentially. #I don't have enough skill to describe the reason because it's not a technical but language issue. Hideki >I don't understand which element makes you say that >section 2 and 3 are all for a 20 block instance > > >Le 27/10/2017 E01:49, Hideki Kato a écrit : >> The 40 block version (2nd instance) first appeared in >> Section 4 in the paper. Section 2 and 3 are all for the 1st >> instance. >> >> Hideki >> >> Xavier Combelle: <39a79a0e-7c7d-2a01-a2ae-573cda8b1...@gmail.com>: >>> Unless I mistake figure 3 shows the plot of supervised learning to >>> reinforcement learning, not 20 bloc/40 block >>> For searching mention of the 20 blocks I search for 20 in the whole >>> paper and did not found any other mention >>> than of the kifu thing. >>> Le 26/10/2017 E15:10, Gian-Carlo Pascutto a écrit : >>>> On 26-10-17 10:55, Xavier Combelle wrote: >>>>> It is just wild guesses based on reasonable arguments but without >>>>> evidence. >>>> David Silver said they used 40 layers for AlphaGo Master. That's more >>>> evidence than there is for the opposite argument that you are trying to >>>> make. The paper certainly doesn't talk about a "small" and a "big" >>> Master. >>>> You seem to be arguing from a bunch of misreadings and >>>> misunderstandings. For example, Figure 3 in the paper shows the Elo >plot >>>> for the 20 block/40 layer version, and it compares to Alpha Go Lee, not >>>> Alpha Go Master. The Alpha Go Master line would be above the flattening >>>> part of the 20 block/40 layer AlphaGo Zero. I guess you missed this >when >>>> you say that they "only mention it to compare on kifu prediction"? >>> _______ >>> 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] November KGS bot tournament
Oh, I didn't notice. S sad. Since AlphaGo, almost all academic organizations have stopped development but, instead, companies and privates are very actively developing these days. Perhaps, many of them don't know this tournament. Hideki Hiroshi Yamashita: <4EDFAD125423416E9D75B2650D2040B1@i3540>: >Hi Nick, > >> this will be the last of the series of KGS bot tournaments. > >Thank you for holding KGS tournament since 2005. >On CGOS, there are always some new comers. >I hope they also enter KGS bot tournament. > >Thanks, >Hiroshi Yamashita > > >- Original Message - >From: "Nick Wedd" <mapr...@gmail.com> >To: <computer-go@computer-go.org> >Sent: Thursday, October 26, 2017 4:43 PM >Subject: [Computer-go] November KGS bot tournament > > >The November KGS bot tournament will be on Sunday, November 5th, starting >at 16:00 UTC and ending by 22:00 UTC. It will use 19x19 boards, with >time limits >of 14 minutes each and very fast Canadian overtime, and komi of 7½. It >will be a Swiss tournament. See http://www.gokgs.com/tournInfo.jsp?id=112 ><http://www.gokgs.com/tournInfo.jsp?id=1116>7 > >Please register by emailing me at mapr...@gmail.com, with the words "KGS >Tournament Registration" in the email title. >With the falling interest in these events since the advent of AlphaGo, it >is likely that this will be the last of the series of KGS bot tournaments. > >Nick >-- >Nick Wedd mapr...@gmail.com > >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Zero is weaker than Master!?
The 40 block version (2nd instance) first appeared in Section 4 in the paper. Section 2 and 3 are all for the 1st instance. Hideki Xavier Combelle: <39a79a0e-7c7d-2a01-a2ae-573cda8b1...@gmail.com>: >Unless I mistake figure 3 shows the plot of supervised learning to >reinforcement learning, not 20 bloc/40 block > >For searching mention of the 20 blocks I search for 20 in the whole >paper and did not found any other mention > >than of the kifu thing. > > >Le 26/10/2017 à 15:10, Gian-Carlo Pascutto a écrit : >> On 26-10-17 10:55, Xavier Combelle wrote: >>> It is just wild guesses based on reasonable arguments but without >>> evidence. >> David Silver said they used 40 layers for AlphaGo Master. That's more >> evidence than there is for the opposite argument that you are trying to >> make. The paper certainly doesn't talk about a "small" and a "big" >Master. >> >> You seem to be arguing from a bunch of misreadings and >> misunderstandings. For example, Figure 3 in the paper shows the Elo plot >> for the 20 block/40 layer version, and it compares to Alpha Go Lee, not >> Alpha Go Master. The Alpha Go Master line would be above the flattening >> part of the 20 block/40 layer AlphaGo Zero. I guess you missed this when >> you say that they "only mention it to compare on kifu prediction"? >> > >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] November KGS bot tournament
The link should be <http://www.gokgs.com/tournEntrants.jsp?id=1127>. Hideki Nick Wedd: <CAEVtG+M-wEEHD-8GLu_QS0KoxW85Svhpf0V3PDGe=Pm=drr...@mail.gmail.com>: >The November KGS bot tournament will be on Sunday, November 5th, starting >at 16:00 UTC and ending by 22:00 UTC. It will use 19x19 boards, with >time limits >of 14 minutes each and very fast Canadian overtime, and komi of 7½. It >will be a Swiss tournament. See http://www.gokgs.com/tournInfo.jsp?id=112 ><http://www.gokgs.com/tournInfo.jsp?id=1116>7 > >Please register by emailing me at mapr...@gmail.com, with the words "KGS >Tournament Registration" in the email title. >With the falling interest in these events since the advent of AlphaGo, it >is likely that this will be the last of the series of KGS bot tournaments. > >Nick -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Zero is weaker than Master!?
Xavier Combelle: <62b977d7-d227-a74d-04b7-0d46db6a7...@gmail.com>: >It is just wild guesses based on reasonable arguments but without >evidence. Yes, of course. Due to not enough info provided by Google. Hideki >Le 26/10/2017 à 07:51, Hideki Kato a écrit : >> You can believe >>> Of what I understand same network architecture imply the same number of >>> block >> but David Silver told AlphaGo Master used 40 layers in >> May. >> http://www.bestchinanews.com/Science-Technology/1 0371.html >> # The paper was submitted in April. >> >> Usually, network "architecture" does not imply the num ber of >> layers whereas "configulation" may do. >> >> Clearly they made 40 layers version first because it's >> called "1st instance" where the 80 layers one is called "2nd >> instance." The 1st was trained 3 days and overtoo k AlphaGo >> Lee. Then they changed to the 2nd. Awaring t his fact, and >> watching the growing curve of the 1st, I g uess 40 layers was >> not enough to reach AlphaGo Master le vel and so they >> doubled the layers. >> >> Hideki >> >> Xavier Combelle: <1550c907-8b96-e4ea-1f5e-2344f394b967 @gmail.com>: >>> As I understand the paper they directly cre ated alphago zero with a 40 >>> block >>> setup. >>> They just made a reduced 20 block setup to co mpare on kifu prediction >>> (as far as I searched in the pa per, it is the only >>> place where they mention the 20 bloc k setup) >>> They specifically mention comparing several ver sion of their software. >>> with various parameter >>> If the number of block was an important parameter I hope they would >>> mention it. >>> Of course they are a lot of things that they try and failed and we will >>> not know about >>> But I have hard time to believe that alphago zero with a 20 block is one >>> of them >>> About the paper, there is no mention of the number of block of master: >>> "AlphaGo Master is the program that defeated top human players by 600 >>> in January, 2017 34 . >>> It was previously unpublished but uses the same neural network >>> architecture, reinforcement >>> learning algorithm, and MCTS algorithm as described in this paper. >>> However, it uses the >>> same handcrafted features and rollouts as AlphaGo Lee >>> and training was initialised by >>> supervised learning from human data." >>> Of what I understand same network architecture imply the same number of >>> block >>> Le 25/10/2017 à 17:58, Xavier Combelle a écrit : >>>> I understand better >>>> Le 25/10/2017 à 04:28, Hideki Kato a écrit : >>>>> Are you thinking the 1st instance could reach Master level >>>>> if giving more training days? >>>>> I don't think so. The performance would be stopping >>>>> improving at 3 days. If not, why they built the 2nd >>>>> instance? >>>>> Best, >>>>> Hideki >>>>> Xavier Combelle: <05c04de1-59c4-8fcd-2dd1-094faabf3...@gmail.com>: >>>>>> How is it a fair comparison if there is only 3 days of training for >>> Zero ? >>>>>> Master had longer training no ? Moreover, Zero has bootstrap problem >>>>>> because at the opposite of Master it don't learn from expert games >>>>>> which means that it is likely to be weaker with little training. >>>>>> Le 24/10/2017 à 20:20, Hideki Kato a écrit : >>>>>>> David Silver told Master used 40 layers network in May. >>>>>>> According to new paper, Master used the same architecture >>>>>>> as Zero. So, Master used 20 blocks ResNet. >>>>>>> The first instance of Zero, 20 blocks ResNet version, is >>>>>>> weaker than Master (after 3 days training). So, with the >>>>>>> same layers (a fair comparison) Zero is weaker than >>>>>>> Master. >>>>>>> Hideki >>>>>> ___ >>>>>> 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Zero is weaker than Master!?
You can believe >Of what I understand same network architecture imply the same number of >block but David Silver told AlphaGo Master used 40 layers in May. http://www.bestchinanews.com/Science-Technology/10371.html # The paper was submitted in April. Usually, network "architecture" does not imply the number of layers whereas "configulation" may do. Clearly they made 40 layers version first because it's called "1st instance" where the 80 layers one is called "2nd instance." The 1st was trained 3 days and overtook AlphaGo Lee. Then they changed to the 2nd. Awaring this fact, and watching the growing curve of the 1st, I guess 40 layers was not enough to reach AlphaGo Master level and so they doubled the layers. Hideki Xavier Combelle: <1550c907-8b96-e4ea-1f5e-2344f394b...@gmail.com>: >As I understand the paper they directly created alphago zero with a 40 >block >setup. > >They just made a reduced 20 block setup to compare on kifu prediction >(as far as I searched in the paper, it is the only >place where they mention the 20 block setup) > >They specifically mention comparing several version of their software. >with various parameter > >If the number of block was an important parameter I hope they would >mention it. > >Of course they are a lot of things that they try and failed and we will >not know about > >But I have hard time to believe that alphago zero with a 20 block is one >of them > >About the paper, there is no mention of the number of block of master: > >"AlphaGo Master is the program that defeated top human players by 600 >in January, 2017 34 . >It was previously unpublished but uses the same neural network >architecture, reinforcement >learning algorithm, and MCTS algorithm as described in this paper. >However, it uses the >same handcrafted features and rollouts as AlphaGo Lee >and training was initialised by >supervised learning from human data." > >Of what I understand same network architecture imply the same number of >block > >Le 25/10/2017 à 17:58, Xavier Combelle a écrit : >> I understand better >> >> >> Le 25/10/2017 à 04:28, Hideki Kato a écrit : >>> Are you thinking the 1st instance could reach Master level >>> if giving more training days? >>> >>> I don't think so. The performance would be stopping >>> improving at 3 days. If not, why they built the 2nd >>> instance? >>> >>> Best, >>> Hideki >>> >>> Xavier Combelle: <05c04de1-59c4-8fcd-2dd1-094faabf3...@gmail.com>: >>>> How is it a fair comparison if there is only 3 days of training for >Zero ? >>>> Master had longer training no ? Moreover, Zero has bootstrap problem >>>> because at the opposite of Master it don't learn from expert games >>>> which means that it is likely to be weaker with little training. >>>> Le 24/10/2017 à 20:20, Hideki Kato a écrit : >>>>> David Silver told Master used 40 layers network in May. >>>>> According to new paper, Master used the same architecture >>>>> as Zero. So, Master used 20 blocks ResNet. >>>>> The first instance of Zero, 20 blocks ResNet version, is >>>>> weaker than Master (after 3 days training). So, with the >>>>> same layers (a fair comparison) Zero is weaker than >>>>> Master. >>>>> Hideki >>>> ___ >>>> 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Zero is weaker than Master!?
Are you thinking the 1st instance could reach Master level if giving more training days? I don't think so. The performance would be stopping improving at 3 days. If not, why they built the 2nd instance? Best, Hideki Xavier Combelle: <05c04de1-59c4-8fcd-2dd1-094faabf3...@gmail.com>: >How is it a fair comparison if there is only 3 days of training for Zero ? >Master had longer training no ? Moreover, Zero has bootstrap problem >because at the opposite of Master it don't learn from expert games >which means that it is likely to be weaker with little training. > > >Le 24/10/2017 à 20:20, Hideki Kato a écrit : >> David Silver told Master used 40 layers network in May. >> According to new paper, Master used the same architecture >> as Zero. So, Master used 20 blocks ResNet. >> >> The first instance of Zero, 20 blocks ResNet version, is >> weaker than Master (after 3 days training). So, with the >> same layers (a fair comparison) Zero is weaker than >> Master. >> >> Hideki > > >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
[Computer-go] Zero is weaker than Master!?
David Silver told Master used 40 layers network in May. According to new paper, Master used the same architecture as Zero. So, Master used 20 blocks ResNet. The first instance of Zero, 20 blocks ResNet version, is weaker than Master (after 3 days training). So, with the same layers (a fair comparison) Zero is weaker than Master. Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Alphago Zero special circumstances
Zero's MCTS code knows "legal" moves (as a part of rules) embed (page 22). Bent four will be solved in practice as she uses Chinese rules. Hideki Dave Dyer: <20171023210343.e65bc31a...@eugeneweb.com>: > >I wonder how alphago-0 treats the menagerie of special positions, such as >bent 4 in the corner, thousand year ko, rotating ko, etc. > >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] AlphaGo Zero
omputer-go.org <mailto:Computer-go@computer-go.org> >>> http://computer-go.org/mailman/listinfo/computer-go >>> <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 >> > >--- >Diese E-Mail wurde von Avast Antivirus-Software auf Viren geprüft. >https://www.avast.com/antivirus > >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] FineArt/JuEZy plays on CGOS!
CGOS 19x19 uses GNU Go v3.7.10 as the anchor, rated 1800. Best, Hideki Nick Wedd: <CAEVtG+MVn+22WC0+Vy0qY6wq=wtfknbjgfqnq3ca7a0h-nt...@mail.gmail.com>: >Hi Magnus, > >Thank you for the information. I don't know how to interpret it. Is there >any relationship between these two lists of Elo ratings? >http://www.yss-aya.com/cgos/19x19/bayes.html >https://en.wikipedia.org/wiki/Go_ranks_and_ratings > >Best, >Nick > >On 17 July 2017 at 12:40, <valky...@phmp.se> wrote: > >> http://www.yss-aya.com/cgos/19x19/cross/JuEYi.html >> >> Best >> Magnus >> ___ >> Computer-go mailing list >> Computer-go@computer-go.org >> http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] purpose of zero-filled feature planes in CNN
AlphaGo's zero plane of the policy network is used as the color feature for the value network (Extended Data Table 2, page 31). These networks share the same architecture so that the value network can be initialized by the policy network before training. Hideki Brian Lee: <CAGCyZsfZyRtA-P+B4G8f+E=to-cu6x2x4vbxaempmqpd4m2...@mail.gmail.com>: >I've been wondering about something I've seen in a few papers (AlphaGo's >paper, Cazenave's resnet policy architecture), which is the presence of an >input plane filled with 0s. > >The input features also typically include a plane of 1s, which makes sense >to me - zero-padding before a convolution means that the 0/1 demarcation >line tells the CNN where the edge of the board is. But as far as I can >tell, a plane of constant 0s should do absolutely nothing. Can anyone >enlighten me? > inline file >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] KGS Bot tournament July
Dear Ingo, Don't worry, the tournament has finished normaly and Zen won. The detail will be announced later by Nick. #The service that update the web page looks broken. Best, Hideki Ingo Althofer: <trinity-77914eb1-98dd-437d-9363-62e71afecf23-1499614872425@3capp-gmx-bs29>: >Hello, > >it seems that the KGS bot tournament did not start, yet. >What is the matter? > >Ingo. >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] NNGS server for Go in Leiden
Dear Ingo and all, It's very sad you are absent... Hiroshi's NNGS server, btw, is running now but looks unstable. We are using my backup server, whose IP is 219.197.124.123 and the port is 9696. Sorry for inconvinient. The 1st round of 9x9 Go will start at two PM local time (GMT+0700), in about one hour. Best, Hideki Ingo Althofer: <trinity-ed96586d-e369-41c9-8d82-6a23fcb348f1-1498894821528@3capp-gmx-bs07>: >Hi Hideki, > >thank you for the information. >Unfortunately, I can not be in Leiden this time. >Please, give my greetings to the other participants, >to the organizers, and also to the participants in >the conference. > >Best regards, Ingo. > >> Gesendet: Samstag, 01. Juli 2017 um 07:11 Uhr >> Von: "Hideki Kato" <hideki_ka...@ybb.ne.jp> >> An: computer-go@computer-go.org >> Betreff: [Computer-go] NNGS server for Go in Leiden ... >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
[Computer-go] NNGS server for Go in Leiden
All, #re-sending because previous post was too long. I'll use Hiroshi's NNGS server for the Go games in Computer Olympiad Leiden 2017. #Thanks, Hiroshi. Its URL is yss-aya.com, port is 9696. Currently planned schedule for 9x9, 13x13 and 19x19 Go follows. You can watch every game via any NNGS client, such as Jago. Time is local (GMT+0200; with DST). #It's five AM July 1st now in Leiden. 7/2 (Sun) 9x9, double round-robin, 4 participants 1000 - 1100 meeting and setup 1100 - 1200 round 1 1400 - 1500 round 2 1500 - 1515 break 1515 - 1615 round 3 7/3 (Mon) 19x19, round-robin, 4 participants 1400 - 1430 meeting and setup 1430 - 1600 round 1 7/4 (Tue) no games 7/5 (Wed) 19x19 cont. 1400 - 1530 round 2 1530 - 1545 break 1545 - 1715 round 3 7/6 (Thu) 13x13, round-robin, 3 participants 1400 - 1430 meeting and setup 1430 - 1530 round 1 1530 - 1545 break 1545 - 1645 round 2 program account entry --- CGI cgi all EZGOezg 9x9 Golois gol 19x19 Julie jul all Zen zen all Best, Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] What was the final score after the counting of AlphaGo-vs-Ke Jie Game #1?
No. With Chinese rules and 7.5 komi, the result is W+0.5. With Japanese rules and 7.5 komi, the result is W+1.5. Because black played the final move in the game (so the number of black stones is one more than white). Hideki Michael Alford: <ba63c6f6-a67e-6a9d-2033-193e4ce29...@aracnet.com>: >Half point is the result using Japanese komi 6.5, Chinese komi is 7.5, >end result is W +1.5 > > >On 5/23/17 12:02 AM, Jim O'Flaherty wrote: >> I have now heard that AlphaGo one by 0.5 points. >> >> >> On Tue, May 23, 2017 at 2:00 AM, Jim O'Flaherty >> <jim.oflaherty...@gmail.com <mailto:jim.oflaherty...@gmail.com>> wrote: >> >> The announcer didn't have her mic on, so I couldn't hear the final >> score announced... >> >> So, what was the final score after the counting of AlphaGo-vs-Ke >> Jie Game #1? >> >> >> >> >> ___ >> Computer-go mailing list >> Computer-go@computer-go.org >> http://computer-go.org/mailman/listinfo/computer-go > inline file >_______ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] mini-max with Policy and Value network
Alvaro Begue:
Re: [Computer-go] mini-max with Policy and Value network
Gian-Carlo Pascutto: <a0b16b16-6591-a195-1f93-93dbe8833...@sjeng.org>: >On 23-05-17 10:51, Hideki Kato wrote: >> (2) The number of possible positions (input of the value net) in >> real games is at least 10^30 (10^170 in theory). If the value >> net can recognize all? L depend on very small difference of >> the placement of stones or liberties. Can we provide necessary >> amount of training data? Have the network enough capacity? >> The answer is almost obvious by the theory of function >> approximation. (ANN is just a non-linear function >> approximator.) > >DCNN clearly have some ability to generalize from learned data and >perform OK even with unseen examples. So I don't find this a very >compelling argument. It's not like Monte Carlo playouts are going to >handle all sequences correctly either. CNN can generalize if global shapes can be built from smaller local shapes. L of a large group is an exception because it's too sensitive for the detail of the position (ie, can be very global). We can't have much expects on such generalization in L By our experiments, value net thinks a group is living if it has a large enough space. That's all. #Actually, this is an opposit. Value net thinks a group is dead if and only if it has short liberties. Some nakade shapes can be solved if outer libeties are almost filled. Additionally, value net frequently thinks false eyes as true, especially on the first lines. (This problem can also be very global and very hard to be solved with no search.) Value net itself cannot manage L correctly but allows so deeper search that this problem is hidden (ie, hard to be known). >Evaluations are heuristic guidance for the search, and a help when the >search terminates in an unresolved position. Having multiple independent >ones improves the accuracy of the heuristic - a basic ensemble. Value net approximates "true" value function of Go very coarsely. Rollouts (MC simulations) fill the detail. This could be a best ensemble. >>(3) CNN cannot learn exclusive-or function due to the ReLU >>activation function, instead of traditional sigmoid (tangent >> hyperbolic). CNN is good at approximating continuous (analog) >> functions but Boolean (digital) ones. > >Are you sure this is correct? Especially if we allow leaky ReLU? Do you know the success of "DEEP" CNN comes from the use of ReLU? Sigmoid easily vanishes gradient while ReLU not. However, ReLU cannot represent sharp edges while sigmoid can. DCNN (with ReLU) approximates functions in a piece-wise-linear style. Hideki ReLU) approximates functions in a piece-wise-linear style. Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] mini-max with Policy and Value network
Gian-Carlo Pascutto: <0357614a-98b8-6949-723e-e1a849c75...@sjeng.org>: >Now, even the original AlphaGo played moves that surprised human pros >and were contrary to established sequences. So where did those come >from? Enough computation power to overcome the low probability? >Synthesized by inference from the (much larger than mine) policy network? Demis Hassabis said in a talk: After the game with Sedol, the team used "adversarial learning" in order to fill the holes in policy net (such as the Sedol's winning move in the game 4). Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] mini-max with Policy and Value network
Erik van der Werf: <cakkggroxqldtkk3vgrernej9diihc_mixxzf75y4wm7grxd...@mail.gmail.com>: >On Tue, May 23, 2017 at 10:51 AM, Hideki Kato <hideki_ka...@ybb.ne.jp> >wrote: > >> Agree. >> >> (1) To solve L, some search is necessary in practice. So, the >> value net cannot solve some of them. >> (2) The number of possible positions (input of the value net) in >> real games is at least 10^30 (10^170 in theory). If the value >> net can recognize all? L depend on very small difference of >> the placement of stones or liberties. Can we provide necessary >> amount of training data? Have the network enough capacity? >> The answer is almost obvious by the theory of function >> approximation. (ANN is just a non-linear function >> approximator.) >> > >A similar argument can be made for natural neural nets, but we know humans >are able to come up with reasonable solutions. I suppose a pure neural net >approach would require some form of recursion, but when combined with a >search, and rolling out the decision process to some sufficiently high >number of max steps, apparently it's not that important.. Also, I suspect >that nearly all positions can only be reached in real games by inferior >moves from both sides. All that may be needed is some crude means to steer >away from chaos (and even if one would start in chaos, humans probably >wouldn't do well either). My argument is for "stand-alone" DCNN. Adding some (top-down?) control to DCNNs could solve this (like human's brain). #I'm not sure about recurrency but maybe necessary. >(3) CNN cannot learn exclusive-or function due to the ReLU >> activation function, instead of traditional sigmoid (tangent >> hyperbolic). CNN is good at approximating continuous (analog) >> functions but Boolean (digital) ones. >> > > >Are you sure about that? I can imagine using two ReLU units to construct a >sigmoid-like step function, so I'd think a multi-layer net should be fine >(just like with ordinary perceptrons). Even if using many layers, it's hard to represent sharp edges by combining ReLUs. (Not impossible but chances are few probably due to so many local traps.) Best, Hideki >Best, >Erik > inline file >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] mini-max with Policy and Value network
istakes without Ko involved, >even >>>> though Ko's drastically increase Leelas fumble chance. >>>> At the same time, Leela is completely and utterly outplaying me on a >>>> strategical level and whenever it manages to not make screwups like the >>>> ones shown I stand no chance at all. Even 3 stones is a serious >challenge >>>> for me then. But those mistakes are common enough to keep me around >even. >>>> >>>> 2017-05-22 17:47 GMT+02:00 Erik van der Werf <erikvanderw...@gmail.com>: >>>> >>>>> On Mon, May 22, 2017 at 3:56 PM, Gian-Carlo Pascutto <g...@sjeng.org> >>>>> wrote: >>>>> >>>>>> On 22-05-17 11:27, Erik van der Werf wrote: >>>>>> > On Mon, May 22, 2017 at 10:08 AM, Gian-Carlo Pascutto <g...@sjeng.org >>>>>> > <mailto:g...@sjeng.org>> wrote: >>>>>> > >>>>>> > ... This heavy pruning >>>>>> > by the policy network OTOH seems to be an issue for me. My >>>>>> program has >>>>>> > big tactical holes. >>>>>> > >>>>>> > >>>>>> > Do you do any hard pruning? My engines (Steenvreter,Magog) always >>>>>> had a >>>>>> > move predictor (a.k.a. policy net), but I never saw the need to do >>>>>> hard >>>>>> > pruning. Steenvreter uses the predictions to set priors, and it is >>>>>> very >>>>>> > selective, but with infinite simulations eventually all potentially >>>>>> > relevant moves will get sampled. >>>>>> >>>>>> With infinite simulations everything is easy :-) >>>>>> >>>>>> In practice moves with, say, a prior below 0.1% aren't going to get >>>>>> searched, and I still regularly see positions where they're the >winning >>>>>> move, especially with tactics on the board. >>>>>> >>>>>> Enforcing the search to be wider without losing playing strength >>>>>> appears >>>>>> to be hard. >>>>>> >>>>>> >>>>> Well, I think that's fundamental; you can't be wide and deep at the >>>>> same time, but at least you can chose an algorithm that (eventually) >>>>> explores all directions. >>>>> >>>>> BTW I'm a bit surprised that you are still able to find 'big tactical >>>>> holes' with Leela now playing as 8d KGS >>>>> >>>>> Best, >>>>> Erik >>>>> >>>>> >>>>> >>>>> ___ >>>>> 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 >>> >> >> > inline file >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] AlphaGo watch parties planned across U.S.
Live will start at 11:00 AM JST (GMT+0900). https://events.google.com/alphago2017/ Hideki Stephen Martindale: <cabmkly+p3aoiwcorzowgykfxrw-ke+r++xfvhgf4xrrts7w...@mail.gmail.com>: >Interesting... but I don't see the links for "watch the games live", only >the schedule. I assume that the page will be updated when the games >actually start... or are about to. > >Yours Faithfully, >*Stephen Martindale* > >+49 162 577 2166 >stephen.c.martind...@gmail.com > >On 22 May 2017 at 09:07, Hideki Kato <hideki_ka...@ybb.ne.jp> wrote: > >> Probably here. >> http://events.google.com/alphago2017/ >> "Below, you can find out the game schedule, watch the games live >> or replay the highlights." >> >> Hideki >> >> Joshua Shriver:
Re: [Computer-go] AlphaGo watch parties planned across U.S.
Three hours earlier, DeepMind twitted: >The Future of Go Summit begins tomorrow. Catch all #AlphaGo17 >matches live from 23-27 May at https://goo.gl/pVjexr Check https://twitter.com/search?q=%23AlphaGo17 ! Hideki Hideki Kato: <59228e17.7176%hideki_ka...@ybb.ne.jp>: >Probably here. >http://events.google.com/alphago2017/ >"Below, you can find out the game schedule, watch the games live >or replay the highlights." > >Hideki > >Joshua Shriver: >
Re: [Computer-go] AlphaGo watch parties planned across U.S.
Probably here. http://events.google.com/alphago2017/ "Below, you can find out the game schedule, watch the games live or replay the highlights." Hideki Joshua Shriver:
Re: [Computer-go] May KGS bot tournament
I was sleeping at the beginning and woked up 3:10 AM, after the 4th round. #The tournament started at 1 AM JST. Hideki Ingo Althofer: <trinity-fb09b52f-df7d-43e9-88fc-1566b2caf541-1494312807584@3capp-gmx-bs13>: >Ui, what happened to Zen in the first four rounds? > >https://www.gokgs.com/tournEntrants.jsp?sort=s=1113 > >Ingo. >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Different Rules
Pawe Morawiecki: <caksbshqxjoqvm9f03xfu8vcboac8aabvpvt0q1wafxpd893...@mail.gmail.com>: >Hideki, > > >> An important difference from actual game is >> the search tree, which is very big in real, long-time setting >> game. One possible interpretation is, Zen read in deep and >> found the (wrong) seki, which would lead W a sure win and so, >> played R18 toward this (again wrong!) winning position. >> >> Looks like DeepZenGo Team just missed a couple of months (weeks?) to train >stronger value network to be able to win the tournament. Michael Redmond 9p >said that DeepZen already plays at the top professional level, particularly >opening and middle game. Congratulations on today's well-deserved win >against Iyama! Thanks. >When would be possible to buy a new DeepZen? Fully depends on the publisher of Tencho-no-Igo, Mynavi. This version will be about one stone weaker on a gaming PC (eight-core Intel with GTX-1080, for example) and two stones or three weaker on a laptop. Best, Hideki >Regards, >Pawel > > > > >> Hideki >> >> Hideki Kato: <58d26196.6952%hideki_ka...@ybb.ne.jp>: >> >We have set komi to 5.5 today. This looks worked fine. >> > >> >The strange yose moves were caused by unknown reason. We are >> >seeking the cause(s). Observed fact: The upper left center >> >three black stones cannot be captured but Zen looks evaluated >> >them as dead. When Zen noticed the truth, horizen effect forced >> >several miserable moves in upper side white territory. Then, >> >upper left white stones together with many short-liberty stones >> >forced the value network misrecognized them as >> >living by seki, because the shape looked seki (for VN) and many >> >moves were required to capture them in rollout. >> > >> >Hideki >> > >> >Pawe Morawiecki: >> ><caksbshogyyn8wk2htv0xczavggem4jj-vpsz_fmqqczq7l8...@mail.gmail.com>: >> >>> >> >>> >> >>> RATHER OFTEN the outcome was a score where both sides thought >> >>> to have won. In the 5.5/7.5 komi example from Go this means that >> >>> outcomes with +6 or +7 points for Black on the board would occur >> >>> often. >> >>> >> >>> >> >>It looks like this issue is serious again was a factor in today's game >> >>against Park 9p. Zen was winning and in the endgame starts giving away >> >>points and the game was reversed. >> >>Hideki, was that the case? >> >> >> >>Too bad it's 6.5 komi as it seems Zen has potential to win both games >:-( >> >> >> >>Regards, >> >>Pawel >> >> >> >> >> >> >> >> >> >>> Of course, this is not welcome for zero-sum games. But it is a hint >> >>> that in reallife scenarios (with non-zero-sum payoffs) Monte Carlo >> >>> heuristics (with their tendency to produce narrow wi0ns) might be >> >>> helpful in finding good compromises. >> >>> >> >>> Ingo. >> >>> ___ >> >>> Computer-go mailing list >> >>> Computer-go@computer-go.org >> >>> http://computer-go.org/mailman/listinfo/computer-go >> >> inline file >> >>___ >> >>Computer-go mailing list >> >>Computer-go@computer-go.org >> >>http://computer-go.org/mailman/listinfo/computer-go >> >-- >> >Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> >> >___ >> >Computer-go mailing list >> >Computer-go@computer-go.org >> >http://computer-go.org/mailman/listinfo/computer-go >> -- >> Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> >> ___ >> Computer-go mailing list >> Computer-go@computer-go.org >> http://computer-go.org/mailman/listinfo/computer-go >> > inline file >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Different Rules
We have set komi to 5.5 today. This looks worked fine. The strange yose moves were caused by unknown reason. We are seeking the cause(s). Observed fact: The upper left center three black stones cannot be captured but Zen looks evaluated them as dead. When Zen noticed the truth, horizen effect forced several miserable moves in upper side white territory. Then, upper left white stones together with many short-liberty stones forced the value network misrecognized them as living by seki, because the shape looked seki (for VN) and many moves were required to capture them in rollout. Hideki Pawe Morawiecki: <caksbshogyyn8wk2htv0xczavggem4jj-vpsz_fmqqczq7l8...@mail.gmail.com>: >> >> >> RATHER OFTEN the outcome was a score where both sides thought >> to have won. In the 5.5/7.5 komi example from Go this means that >> outcomes with +6 or +7 points for Black on the board would occur >> often. >> >> >It looks like this issue is serious again was a factor in today's game >against Park 9p. Zen was winning and in the endgame starts giving away >points and the game was reversed. >Hideki, was that the case? > >Too bad it's 6.5 komi as it seems Zen has potential to win both games :-( > >Regards, >Pawel > > > > >> Of course, this is not welcome for zero-sum games. But it is a hint >> that in reallife scenarios (with non-zero-sum payoffs) Monte Carlo >> heuristics (with their tendency to produce narrow wi0ns) might be >> helpful in finding good compromises. >> >> Ingo. >> ___ >> Computer-go mailing list >> Computer-go@computer-go.org >> http://computer-go.org/mailman/listinfo/computer-go > inline file >_______ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Zen lost to Mi Yu Ting
The value network has been trained with Chinese rules and 7.5 pts komi. Using this for Japanese and 6.5, there will be some error in close games. We knew this issue and thought such chances would be so small that postponed correcting (not so easy). Best, Hideki Pawe Morawiecki: <caksbshpvd34hvjt-b+x73rdpg5-4wsxoezykbheslprewci...@mail.gmail.com>: >Hi, > >After an interesting game DeepZen lost to Mi Yu Ting. >Here you can replay the complete game: >http://duiyi.sina.com.cn/gibo_new/live/viewer.asp?sno=13 > >According to pro experts, Zen fought really well, but it seems there is >still some issue how Zen (mis)evaluates its chances. At one point it showed >84% chance of winning (in the endgame), whereas it was already quite clear >Zen is little behind (2-3 points). > >Regards, >Pawel > inline file >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] UEC cup 1st day result
Dear Ingo, Fine Art is developed by a team in Tencent, a big IT company at Beijing, China. Fine Art has been running on Fox-Go, a Chinese internet Go site, where many strong pros in China, Korea and Japan are playing, including Ka Jie and Yuta Iyama. Fine Art has beaten pros at about 90% in fast (30 seconds byoyomi) games and about 70% in 1 minute ones. Moreover, the latest, UEC Cup version is much stronger than this! We will use the UEC Cup verion of Zen (v14.3) for the World Go Championship. If necessary and possible, we might add some improvements, though. #V14.3 is running on CGOS 19x19 now. Using a 3 GHz ivy-bridge core and a GTX-1060 gpu, almost no lose on CGOS. See http://www.yss-aya.com/cgos/19x19/standings.html We used v14.3 on a dual-Xeon server with 4 nVidia Titan-X (Pascal) gpus yesterday but easily beaten by Fine Art. Rayn is a combination of Ray and Rn. Rn is a DCNN enhanced (by Matsuzaki-san) version of an open source strong program Ray. Rn is also running on CGOS (using 4 cores and 1 gpu). Hideki Ingo Althofer: <trinity-01ddc2c6-8cfb-40a6-adc5-b677266ed2e2-1489854290378@3capp-gmx-bs21>: >Dear Hiroshi, > >thank you for your posting. > >Can you say something more on "Fine Art"? >From which country is it? Who is Tencent? > >Is the "UEC-Zen" version similar to the one that will play >in the World Championships next week? > >Who is the team behind Rayn? > >Questions over questions ... >exciting times... > >Cheers, Ingo. > > >> Gesendet: Samstag, 18. März 2017 um 16:27 Uhr >> Von: "Hiroshi Yamashita" <y...@bd.mbn.or.jp> >> An: computer-go@computer-go.org >> Betreff: [Computer-go] UEC cup 1st day result >> >> Today, 30 participanst played, and >> >> 1st Fine Art 7-0 >> 2nd Zen6-1 lost aaginst FineArt >> 3rd CrazyStone 6-1 >> 4th Rayn >> 5th Aya5-2 >> 6th AQ >> 7th CGI >> 8th Julie >> 9th DolBaram >> >> Top 16 program will play tommorow tournament. >> >> Remi's post is useful. >> https://twitter.com/Remi_Coulom/status/843115954473070593 >> >> There was a lot of media this year. Maybe because of "Fine Art" >> (very strong program made by Tencent) >> >> Thanks, >> Hiroshi Yamashita >> Gesendet: Samstag, 18. März 2017 um 16:27 Uhr >> Von: "Hiroshi Yamashita" <y...@bd.mbn.or.jp> >> An: computer-go@computer-go.org >> Betreff: [Computer-go] UEC cup 1st day result >> >> Today, 30 participanst played, and >> >> 1st Fine Art 7-0 >> 2nd Zen6-1 lost aaginst FineArt >> 3rd CrazyStone 6-1 >> 4th Rayn >> 5th Aya5-2 >> 6th AQ >> 7th CGI >> 8th Julie >> 9th DolBaram >> >> Top 16 program will play tommorow tournament. >> >> Remi's post is useful. >> https://twitter.com/Remi_Coulom/status/843115954473070593 >> >> There was a lot of media this year. Maybe because of "Fine Art" >> (very strong program made by Tencent) >> >> Thanks, >> Hiroshi Yamashita >> >> ___ >> 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] UEC cup 1st day result
Dear Ingo, Fine Art is developed by a team in Tencent, a big IT company at Beijing, China. Fine Art has been running on Fox-Go, a Chinese internet Go site, where many strong pros in China, Korea and Japan are playing, including Ka Jie and Yuta Iyama. Fine Art has beaten pros at about 90% in fast (30 seconds byoyomi) games and about 70% in 1 minute ones. Moreover, the latest, UEC Cup version is much stronger than this! We will use the UEC Cup verion of Zen (v14.3) for the World Go Championship. If necessary and possible, we might add some improvements, though. #V14.3 is running on CGOS 19x19 now. Using a 3 GHz ivy-bridge core and a GTX-1060 gpu, almost no lose on CGOS. See http://www.yss-aya.com/cgos/19x19/standings.html We used v14.3 on a dual-Xeon server with 4 nVidia Titan-X (Pascal) gpus yesterday but easily beaten by Fine Art. Rayn is a combination of Ray and Rn. Rn is a DCNN enhanced (by Matsuzaki-san) version of an open source strong program Ray. Rn is also running on CGOS (using 4 cores and 1 gpu). Hideki Ingo Althofer: <trinity-01ddc2c6-8cfb-40a6-adc5-b677266ed2e2-1489854290378@3capp-gmx-bs21>: >Dear Hiroshi, > >thank you for your posting. > >Can you say something more on "Fine Art"? >From which country is it? Who is Tencent? > >Is the "UEC-Zen" version similar to the one that will play >in the World Championships next week? > >Who is the team behind Rayn? > >Questions over questions ... >exciting times... > >Cheers, Ingo. > > >> Gesendet: Samstag, 18. März 2017 um 16:27 Uhr >> Von: "Hiroshi Yamashita" <y...@bd.mbn.or.jp> >> An: computer-go@computer-go.org >> Betreff: [Computer-go] UEC cup 1st day result >> >> Today, 30 participanst played, and >> >> 1st Fine Art 7-0 >> 2nd Zen6-1 lost aaginst FineArt >> 3rd CrazyStone 6-1 >> 4th Rayn >> 5th Aya5-2 >> 6th AQ >> 7th CGI >> 8th Julie >> 9th DolBaram >> >> Top 16 program will play tommorow tournament. >> >> Remi's post is useful. >> https://twitter.com/Remi_Coulom/status/843115954473070593 >> >> There was a lot of media this year. Maybe because of "Fine Art" >> (very strong program made by Tencent) >> >> Thanks, >> Hiroshi Yamashita >> >> ___ >> 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] UEC cup 1st day result
Dear Ingo, Fine Art is developed by a team in Tencent, a big IT company at Beijing, China. Fine Art has been running on Fox-Go, a Chinese internet Go site, where many strong pros in China, Korea and Japan are playing, including Ka Jie and Yuta Iyama. Fine Art has beaten pros at about 90% in fast (30 seconds byoyomi) games and about 70% in 1 minute ones. Moreover, the latest, UEC Cup version is much stronger than this! We will use the UEC Cup verion of Zen (v14.3) for the World Go Championship. If necessary and possible, we might add some improvements, though. #V14.3 is running on CGOS 19x19 now. Using a 3 GHz ivy-bridge core and a GTX-1060 gpu, almost no lose on CGOS. See http://www.yss-aya.com/cgos/19x19/standings.html We used v14.3 on a dual-Xeon server with 4 nVidia Titan-X (Pascal) gpus yesterday but easily beaten by Fine Art. Rayn is a combination of Ray and Rn. Rn is a DCNN enhanced (by Matsuzaki-san) version of an open source strong program Ray. Rn is also running on CGOS (using 4 cores and 1 gpu). Hideki Ingo Althofer: <trinity-01ddc2c6-8cfb-40a6-adc5-b677266ed2e2-1489854290378@3capp-gmx-bs21>: >Dear Hiroshi, > >thank you for your posting. > >Can you say something more on "Fine Art"? >From which country is it? Who is Tencent? > >Is the "UEC-Zen" version similar to the one that will play >in the World Championships next week? > >Who is the team behind Rayn? > >Questions over questions ... >exciting times... > >Cheers, Ingo. > > >> Gesendet: Samstag, 18. März 2017 um 16:27 Uhr >> Von: "Hiroshi Yamashita" <y...@bd.mbn.or.jp> >> An: computer-go@computer-go.org >> Betreff: [Computer-go] UEC cup 1st day result >> >> Today, 30 participanst played, and >> >> 1st Fine Art 7-0 >> 2nd Zen6-1 lost aaginst FineArt >> 3rd CrazyStone 6-1 >> 4th Rayn >> 5th Aya5-2 >> 6th AQ >> 7th CGI >> 8th Julie >> 9th DolBaram >> >> Top 16 program will play tommorow tournament. >> >> Remi's post is useful. >> https://twitter.com/Remi_Coulom/status/843115954473070593 >> >> There was a lot of media this year. Maybe because of "Fine Art" >> (very strong program made by Tencent) >> >> Thanks, >> Hiroshi Yamashita >> >> ___ >> 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] UEC Cup broadcast info
Individual games can be watched by logging-in the NNGS server using some NNGS client program such as Jago. See official page http://www.computer-go.jp/uec/public_html/eng/network.shtml for the name of the server, which is global this year. NB: Please leave enough bandwidth for the entrants. Hideki Hideki Kato: <58c9dab7.6916%hideki_ka...@ybb.ne.jp>: >Dear all, > >The 10th UEC will be held on this weekend. The final will be >live broadcast on the internet by Go-remium (in Japanese; free >but registration required). The live will start 3/19 2:30 PM >JST (GMT/UTC+9). >#I don't remember the procedure to register, sorry. >http://www.computer-go.jp/uec/public_html/eng/index.shtml >https://www.igoshogi.net/igopremium/live/live_info.html?live_id=c8c9aff640925b4ac0697ec380fca54b > >Also, Ohashi 6p will have a program "Spaceman de Go" starting >at 3/18 8 PM JST (after the preliminary round) on Niconico >movie. Rina Fujisawa 3p will come! >http://live.nicovideo.jp/watch/lv291350901 > >Hideki >-- >Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Congratulations to Zen19X!
Thank you for the tournament and the report. I believe the number of players was 4, rather than 6, and "2016 Annual KGS Bot Championship" should be 2017. Hideki Nick Wedd: <CAEVtG+M0+=cKb_E=oip0hm8_kuyxgkwutwvneaqq4cqej0+...@mail.gmail.com>: >Congratulations to Zen19X, winner of the Spring Slow KGS bot tournament! > >My report is now at http://www.weddslist.com/kgs/past/S17.1/index.html >As always, I will welcome your comments and corrections. > >I apologise for the very late appearance of my report. At first I put it >off because the KGS software I use to build the crosstable had stopped >working, and I hoped that it would be repaired (it is still broken). Then I >was overtaken by other commitments for a few days. > >Nick -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
[Computer-go] UEC Cup broadcast info
Dear all, The 10th UEC will be held on this weekend. The final will be live broadcast on the internet by Go-remium (in Japanese; free but registration required). The live will start 3/19 2:30 PM JST (GMT/UTC+9). #I don't remember the procedure to register, sorry. http://www.computer-go.jp/uec/public_html/eng/index.shtml https://www.igoshogi.net/igopremium/live/live_info.html?live_id=c8c9aff640925b4ac0697ec380fca54b Also, Ohashi 6p will have a program "Spaceman de Go" starting at 3/18 8 PM JST (after the preliminary round) on Niconico movie. Rina Fujisawa 3p will come! http://live.nicovideo.jp/watch/lv291350901 Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Question: Time Table World Championships
Dear Ingo, English official page has the info. http://www.worldgochampionship.net/english/ Hideki Ingo Althofer: <trinity-4ec0963b-d8ae-4f58-9bd0-8ae5d7bf24ef-1489046933442@3capp-gmx-bs79>: >Hi, > >from March 21 to 23, the Go World Championships will >take place: one human participant each from Japan, South-Korea >and China PLUS bot Deep Zen. > >Does someone know already details on the time table >and the online venues where the games are transmitted >or mirrored? > >Thx in advance, Ingo. >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] March KGS bot tournament - slow
Dear Nick, It was caused by a bug around managing value network. Best, Hideki Nick Wedd: <caevtg+no8grcdqdp0kajfaxcgufk+uigagfn0us2y6vx74p...@mail.gmail.com>: >There have been some interesting games in this event. Zen19X has just lost >a game to JulieBot, by passing when the game was over but JulieBot made a >sente move inside its territory. > >Unfortunately, I have not been writing my usual report and updating it >after each round, as there is a problem with the KGS server, that hinders >user access to completed games, and prevents the creation of a web page for >each completed round. Also unfortunately, I shall be busy with other >activities for the next few days, and probably won't be able to write my >report until Monday. > >Nick > >On 27 February 2017 at 12:06, Nick Wedd <mapr...@gmail.com> wrote: > >> The March KGS bot tournament will start on Sunday, March 5th, starting at >> 22:00 UTC and end by 14:00 UTC on Wednesday 8th. It will use 19x19 >> boards, with time limits of 235 minutes each plus fast Canadian overtime, >> and komi of 7½. That's four hours each, so each round will take eight >> hours. >> >> It will be a Swiss tournament, with eight rounds. See >*https://www.gokgs.com/tournInfo.jsp?id=1100 >> <https://www.gokgs.com/tournInfo.jsp?id=1100>* >> >> Please register by emailing me at mapr...@gmail.com, with the words "KGS >> Tournament Registration" in the email title. >> >> Nick >> -- >> Nick Wedd mapr...@gmail.com >> -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] How is zen so strong on CGOS?
Dear Detlef, Please don't forget the rollout, which is still important. Zen's rollout is much better than others' so far. This could be the main reason of your question. The value network introduced by the Nature paper is not perfect, rather represents very rough (coarse) approximation of the value function of 19x19 Go. The number of possible positions is 10^170 and is greater than 10^30 in practice, I guess. The order of the number of the freedom (and the training samples) of the value network is 10^7 which is much much smaller than the number of freedom being necessary to represent the value function. So, the value network only approximate rough, smooth, low-order part of the function. Thus, the detal, high-order part of the function must be represented by the rollout. Best, Hideki Detlef Schmicker: <d8a14055-2f4e-7e99-d9ba-2c79a34af...@physik.de>: >Hi, > >I'd like to start a discussion on what zen might do being so strong on >CGOS with only one core and no graphic card :) > >The version actual playing (and therefore best comparable to the >programs actual playing) is > >Rank NameElo + - Games >12 Zen-13.5-1c0g 363948 48 569 >16 Aya792p2v2cn50_12t 355846 46 476 >24 Rn.3.3-4c 345934 34 918 >26 No335-4.5-gpu-4c3438115 115 112 >29 CGI1407_1_475_7c339973 73 142 >53 CrazyStone-0002 3239114 114 141 >54 NG-05 323567 67 228 > > > >No335 is probably Hirabot >NG is oakfoam >from crazystone I dont know, if this version uses gpu >Rn is probably ray > >All but Zen use gpu and so have probably about 300 policy network calls >per second. > >Zen does this on cpu and I got Hideki to let me know, that this can do >only about 20 calls per second :) > >It might be that zen uses cnn only in the upper nodes and relies on the >faster node generation zen used before cnn for deeper nodes? > >I can not believe, that the policy network is so much stronger than the >ones of all the others?! > > >Any ideas > > >Detlef >_______ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] ADMIN: Maintenance ?!?!
Dear Ingo, Ingo Althofer: <trinity-d23b0c78-589b-4bca-99b2-eec54993d70a-1483420625560@3capp-gmx-bs07>: >Please, no maintenance right now. >The days in computer go are just so hot. > >The team of "Master" offered 14,000 Dollars >for any being beating the bot on January 02 or 03. No, who offered the prize is not the team but Koriki 9p and FoxGo. Hideki >We need the possibility to discuss this topic in realtime. > >Thx, ingo. -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Poll: Scientific Breakthrough of the Year 2016
Now it getting closer. (1) Human embryos in a disc 16 % (2) Ripples in space time 14 % (3) AI in games (AlphaGo) 13% (4) Pocket sized DNA sequencers 8 % (5) The purge that refreshes 7% (5) Custom designed proteins 7 % Hideki Ingo Althofer: <trinity-d4401556-343d-4b1f-aea0-42a19fd5268f-1480514806732@3capp-gmx-bs01>: >Hi, >I just learned that the magazine "Science" is making a poll >on "Scientific Breakthrough of the Year 2016". Voters may >chose amongst 15 proposals. Voting closes on December 04. > >Currently the following 5 subjects have top votes. > >(1) Human embryos in a disc 17 % >(2) Ripples in space time 15 % >(3) AI in games (AlphaGo) 9 % >(4) Pocket sized DNA sequencers 8 % >(5) Custom designed proteins 7 % > >http://www.sciencemag.org/news/2016/11/vote-your-scientific-breakthrough-year > >Feel free to participate. >No registration needed. > >Ingo. >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] World Go Championship
Some kind of invitation, yes. Top players from countries and AI community. They announced that they will continue this tournament at least three years. So, if there will be many strong programs, all (but one from a country) will be invited, I guess. Hideki valky...@phmp.se: <47dc43d9bcbbac1ff0e49729dfd4c...@phmp.se>: >Could be, but the next UEC Cup will be last... so maybe this will be >some kind of invitational event. > >Magnus > >On 2016-11-29 15:52, Michael Markefka wrote: >> Just a wild guess, but I assume they'll go for the latest winner of >> the UEC Cup as far as AI entrants are concerned. >> >> On Tue, Nov 29, 2016 at 3:46 PM, "Ingo Althöfer" >> <3-hirn-ver...@gmx.de> wrote: >> >>> Hi Hideki, >>> >>> that sounds very interesting. >>> >>>> Nihon Kiin created a new Go tournament, "World Go Championship", >>> which >>>> will be held in March 21st to 23rd, in Osaka, Japan. >>>> >>>> Top three professional players from Japan, China and Korea and one >>>> Computer Go program will attend. >>> >>> Big questions: >>> * Where can computer programmers apply on behalf of their bots? >>> * Will there be a qualifier tournament for bots? >>> >>> Looking forward to this event with big eyes, >>> Ingo. >>> >>> ___ >>> Computer-go mailing list >>> Computer-go@computer-go.org >>> http://computer-go.org/mailman/listinfo/computer-go [1] >> >> >> >> Links: >> -- >> [1] 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] World Go Championship
Dear Ingo, For this time, Nihon kiin recommended (and chosen) Zen for the seat probably by the right of the organizer. #Nihon kiin first asked AlphaGo but declined due to its schedule, an article reported. Hideki >Hi Hideki, > >that sounds very interesting. > >> Nihon Kiin created a new Go tournament, "World Go Championship", which >> will be held in March 21st to 23rd, in Osaka, Japan. >> >> Top three professional players from Japan, China and Korea and one >> Computer Go program will attend. > >Big questions: >* Where can computer programmers apply on behalf of their bots? >* Will there be a qualifier tournament for bots? > >Looking forward to this event with big eyes, >Ingo. >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
[Computer-go] World Go Championship
Hi all, Nihon Kiin created a new Go tournament, "World Go Championship", which will be held in March 21st to 23rd, in Osaka, Japan. Top three professional players from Japan, China and Korea and one Computer Go program will attend. Yuta Iyama six-crown and Zen from Japan. China and Korea are in cordination. The prize money is 30 million yen. http://www.nihonkiin.or.jp.e.qs.hp.transer.com/news/release/ai.html Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
[Computer-go] Auto Go game recorder
Hello everybody, Chizu Kobayashi 6p is seeking automatic Go game recorders. Does anyone know about that? An application for mobilephones is the best but any system is appriciated. Best, Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Deep Zen vs Cho Chikun -- Round 3
You're welcome. I've finished the game. More Human vs Computer games at http://computer-go.info/h-c/. Hideki Xavier Combelle: <d21c12c5-4c1c-ea51-c34a-8d35a9174...@gmail.com>: >thanks a lot > > >Le 24/11/2016 à 12:31, Hideki Kato a écrit : >> Ah, yes. Maybe he didn't push "resign". >> >> Hideki >> >> Xavier Combelle: <1d25b061-1f25-497e-c270-ee2040602...@gmail.com>: >>> Hi Hideki >>> Sorry in kgs, the game has no result in kgs, did it ended with pro >>> resignation ? >>> Le 23/11/2016 à 23:03, Hideki Kato a écrit : >>>> Thanks David. >>>> It's now. >>>> In the same afternoon, Zen vs Yonil Ha 6p was played on KGS as a part >>>> of Neyagawa Igo Shogi Festival in Neyagawa city, Osaka, Japan. >>>> (Zen19X vs neyagawa. The time was set 4 hours to avoid KGS's time >>>> control and actually a move was played in 30s). >>>> This Zen ran on a dual Xeon server with one nVidia GTX-1080 in my >>>> room. I ran seven threads. This shows recent Zen on a PC with a >>>> highend GPU is enough to beat pro at short-time settings. >>>> Also, Zen on a dual-core laptop (ThinkPad X250; Core i7 5600U@2.6 GHz) >>>> beat a pro a few times in personal trials (also fast games). >>>> Hideki >>>> David Fotland: <06c901d245a8$d41405f0$7c3c11d0$@smart-games.com>: >>>>> Congratulations to Zen for playing so well against a strong pro. It >>> won't >>>>> be long until anyone can get a pro strength go program that runs on >>> their >>>>> ordinary PC. >>>>> David >>> ___ >>> 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Deep Zen vs Cho Chikun -- Round 3
Ah, yes. Maybe he didn't push "resign". Hideki Xavier Combelle: <1d25b061-1f25-497e-c270-ee2040602...@gmail.com>: >Hi Hideki > >Sorry in kgs, the game has no result in kgs, did it ended with pro >resignation ? > > >Le 23/11/2016 à 23:03, Hideki Kato a écrit : >> Thanks David. >> >> It's now. >> >> In the same afternoon, Zen vs Yonil Ha 6p was played on KGS as a part >> of Neyagawa Igo Shogi Festival in Neyagawa city, Osaka, Japan. >> (Zen19X vs neyagawa. The time was set 4 hours to avoid KGS's time >> control and actually a move was played in 30s). >> >> This Zen ran on a dual Xeon server with one nVidia GTX-1080 in my >> room. I ran seven threads. This shows recent Zen on a PC with a >> highend GPU is enough to beat pro at short-time settings. >> >> Also, Zen on a dual-core laptop (ThinkPad X250; Core i7 5600U@2.6 GHz) >> beat a pro a few times in personal trials (also fast games). >> >> Hideki >> >> David Fotland: <06c901d245a8$d41405f0$7c3c11d0$@smart-games.com>: >>> Congratulations to Zen for playing so well against a strong pro. It >won't >>> be long until anyone can get a pro strength go program that runs on >their >>> ordinary PC. >>> >>> David > >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Deep Zen vs Cho Chikun -- Round 3
Unfortunately, in Japanese software market, unlike Western, patches are provided almost only for bug fixes, not updating. Hideki Michael Markefka:
Re: [Computer-go] Deep Zen vs Cho Chikun -- Round 3
Thanks David. It's now. In the same afternoon, Zen vs Yonil Ha 6p was played on KGS as a part of Neyagawa Igo Shogi Festival in Neyagawa city, Osaka, Japan. (Zen19X vs neyagawa. The time was set 4 hours to avoid KGS's time control and actually a move was played in 30s). This Zen ran on a dual Xeon server with one nVidia GTX-1080 in my room. I ran seven threads. This shows recent Zen on a PC with a highend GPU is enough to beat pro at short-time settings. Also, Zen on a dual-core laptop (ThinkPad X250; Core i7 5600U@2.6 GHz) beat a pro a few times in personal trials (also fast games). Hideki David Fotland: <06c901d245a8$d41405f0$7c3c11d0$@smart-games.com>: >Congratulations to Zen for playing so well against a strong pro. It won't >be long until anyone can get a pro strength go program that runs on their >ordinary PC. > >David -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Deep Zen vs Cho Chikun -- Round 1
>Hi all, > >the match was in German TV (ZDF) today, in an 8 second snipet. > >It included a scene from the press conference: >http://www.dgob.de/yabbse/index.php?action=dlattach;topic=6322.0;attach=5391;image > >I recognise Cho Chikun in the center and Hideki Kato, second from left. >Who are the other three gentlemen? From left to right, Nobuo Kawakami (Chairman of Dwango), Me (Representative of Team DeepZen), Chikun Cho 9p & honor Meijin, Hiroshi Yamashiro 9p (Vice President of Nihon-kiin), Hiroaki Dan (President of Nihon-kiin). Best, Hideki >Cheers, Ingo. > >PS. A photo showing Angela Merkel pressing her thumbs: >http://www.wunderweib.de/assets/field/image/100/merkel7.jpg >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
[Computer-go] English commentary available (Re: Zen vs Cho Chikun)
These games will have English commentary by Michael Redmond 9p and Antti Tormanen 1p. http://live.nicovideo.jp/watch/lv281949598 http://live.nicovideo.jp/watch/lv281949974 http://live.nicovideo.jp/watch/lv281950196 Best, Hideki Hideki Kato: <58232219.6505%hideki_ka...@ybb.ne.jp>: >Thanks Ingo, > >Anyone will be able to watch the games on Niconico video. > >Set the first language of your web browser to English and access >http://www.nicovideo.jp/. You will be able to register (free) and >will be able to watch the games. > >The links to the games are >http://live.nicovideo.jp/watch/lv280070012 (11/19, 12:30 PM JST), >http://live.nicovideo.jp/watch/lv280070055 (11/20, ditto) and >http://live.nicovideo.jp/watch/lv280070087 (11/23, ditto). >You can set (or reserve) "time shift" viewing, which allows you to >watch later. > >Rules: Japanese rules, 6.5 pts komi, 2 hours + 3 x 60 seconds >byoyomi. All three games will be played. > >Hardware: >CPU: 2 x Intel Xeon E5-2699v4 (44 cores/2.2 GHz) >GPU: 4 x nVidia Titan X (Pascal) >RAM: 128GB > >Best, Hideki > >Ingo Althofer: ><trinity-7a1b490a-5533-4ab9-a2b5-e252ea8e4c7d-1478695603496@3capp-gmx-bs75>: >>Hi everybody, > >> > >>there was a press conference by the Zen team, announcing a three >>games > >>match between Zen and human pro hero Cho Chiukum with games to > >>be played on November 19, 20, and 23. > >>Unfortunately for me, the announcement was in Japanese language which > >>I do not understand. > >> > >>For me (and likely for many in the Western world) it would be nice > >>if for instance Hideki Kato could explain what is to come. > >> > >>Thanks in advance, > >>Ingo. > >>___ > >>Computer-go mailing list > >>Computer-go@computer-go.org > >>http://computer-go.org/mailman/listinfo/computer-go >-- >Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Zen vs Cho Chikun
Thanks Ingo, Anyone will be able to watch the games on Niconico video. Set the first language of your web browser to English and access http://www.nicovideo.jp/. You will be able to register (free) and will be able to watch the games. The links to the games are http://live.nicovideo.jp/watch/lv280070012 (11/19, 12:30 PM JST), http://live.nicovideo.jp/watch/lv280070055 (11/20, ditto) and http://live.nicovideo.jp/watch/lv280070087 (11/23, ditto). You can set (or reserve) "time shift" viewing, which allows you to watch later. Rules: Japanese rules, 6.5 pts komi, 2 hours + 3 x 60 seconds byoyomi. All three games will be played. Hardware: CPU: 2 x Intel Xeon E5-2699v4 (44 cores/2.2 GHz) GPU: 4 x nVidia Titan X (Pascal) RAM: 128GB Best, Hideki Ingo Althofer: <trinity-7a1b490a-5533-4ab9-a2b5-e252ea8e4c7d-1478695603496@3capp-gmx-bs75>: >Hi everybody, > >there was a press conference by the Zen team, announcing a three >games >match between Zen and human pro hero Cho Chiukum with games to >be played on November 19, 20, and 23. >Unfortunately for me, the announcement was in Japanese language which >I do not understand. > >For me (and likely for many in the Western world) it would be nice >if for instance Hideki Kato could explain what is to come. > >Thanks in advance, >Ingo. >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
[Computer-go] Chun-Hsun Chou 9p vs Yi-Min Hsieh 6p and DarkForest on ICIRA 2016 Tokyo
Chun-Hsun Chou 9p (W) vs Yi-Min Hsieh 6p and DarkForest (B) was played as a part of the events of ICIRA 2016 Tokyo on August 24. Probably and at least :) Ingo will have some interest. Due to a connection trouble, the game was supended (unfinished) but Black team was clearly ahead (by Ohashi 6p and a strong amateur). This was the second game of the team (the first was played at WCCI 2016 Vancouver). Hsieh 6p looks advanced/studied a lot. She told that she clearly could do better on reading (and played by herself) in certain positions but DarkForest had some advantage in global evaluation (and she picked up one from DF's suggested moves) on early stages (or static positions). Unfortunatelly, the game record (http://oase.nutn.edu.tw/ICIRA2016/result.htm) includes nothing about what moves DarkForest suggested in the game but maybe the developers have some. #Those suggestions were told her by voice of a robot or showed up as an ordered list of moves, ie, with no histgram of scores. Best, Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Zen19K: a KGS-9-dan in the making
Version 12.1 does not run on my hardware. It uses 4 Titan X GPUs. Hideki Petr Baudis: <20160829222749.gg6...@machine.or.cz>: >On Sat, Aug 27, 2016 at 02:04:39PM +0200, "Ingo Althöfer" wrote: >> the cooperation of the Zen team and Dwango provides >> first fat harvest: Zen19K is on the path to become >> the first bot with a 9-dan rating on KGS: >> http://www.gokgs.com/gameArchives.jsp?user=zen19k >> >> Observe: this is not the Zen that will play against >> Luks Kraemer in the codecentric Challenge (from tomorrow on). > >Silly question - why not? (Assuming this seems to be the strongest Zen >so far.) > >Thanks, > > Petr Baudis >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Missing results for Computer Go events
Nick Wedd: <caevtg+mtq7makrw+m-xvqjdozkku+nu+n5dzbhge5onbclt...@mail.gmail.com>: >I have not been able to find any results for three Computer Go events from >this year: > >4th Densei-Sen March 23 >http://entcog.c.ooco.jp/entcog/densei/densei4_En/ Although above English page has no results yet, Japanese page (http://entcog.c.ooco.jp/entcog/densei/densei4/) does have. Kobayashi 9p won the first game (vs. darkforest) and lost the second (vs. Zen). The record of the first game is available at http://entcog.c.ooco.jp/entcog/densei/densei4/densei4_1.sgf and the second at http://entcog.c.ooco.jp/entcog/densei/densei4/densei4_2.sgf . >Korea Baduk Association June or July >http://english.baduk.or.kr/ I believe it was canceled due to the AlphaGo shock. >FuzzIEEE 2016 July 25-29 >http://www.wcci2016.org/ http://oase.nutn.edu.tw/WCCI2016/result.htm Best, Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] cgos <--> kgs rating
GNU Go has 1800 Elo on cgos and is ranked from 8 to 6 kyu on KGS. Zen v11.8 has 3420 on cgos using 2 cores and 8-core version is ranked 8d (blitz) on KGS. The difference must be about 200 Elo. Probably Aya (and HiraBot?) also have two numbers. #Note that the time settings are different for cgos and kgs. Hideki Detlef Schmicker: <57716ada.9060...@physik.de>: >-BEGIN PGP SIGNED MESSAGE- >Hash: SHA1 > >Hay, > >I wonder if somebody has the same program (with the same settings) >rated on cgos 19x19 and kgs? > >I am still fighting with resigning in the case of value-network and >playouts disagree, so I can not run oakfoam on kgs, but would like to >have a strength hint :) > > >Thanks Detlef >-BEGIN PGP SIGNATURE- >Version: GnuPG v2.0.22 (GNU/Linux) > >iQIcBAEBAgAGBQJXcWraAAoJEInWdHg+Znf4vFgP/18AbsZpOH9mjjE4j30j4WZm >R3MJFQW/519IczwBNkps80Q1Gl0K7LYJ6ss7Ok0FUmqVB+3V4+ZXoMLp5wD637VO >nRTXCvdx+xUyyR+fPgj7PEtGCr+hNsbvDlwxqsqdX5laK+L5NanvPMIWz0ZMCFlV >7CUHweIwac2f56l679Urf+wqBXysQiGcOHbXmcWlUVlJ6Og4xKfCJM38Gx2gUyfK >+yPecX6c84t1qe/h9rw2E9tkP9KHM0rpyTsJGLvVh2ia8mmkdP1KZeeFd7sM9pQV >vClV53JJdU//iCqi5/XB2Sn4YTgElk4GjTsb5pC7EmG1QM7+PhU6v7mfYll/4XV1 >xGeLuFRU+eqQvkd0yW+x/GRlgTfbo212nNwuHWNpGe9InCwFEIvcNNe2lPo6B+sD >ldK13nrYdkYVTr4Wa0+WpQaEewv0gA/9nQcjTntKikVcvDwYMn/OwBvukzSb12e3 >yTvpy6Yx41EL4+D8bpI373/1MvLnbLAKssEYSNTKT2o/1gtBqVF40zXya+lnPb+m >s8TOOxWbS4EktrLIbwpD+xzpb2O91TfPQO1/xLA+12g/SMIvW4OB6ghAiPP8WHsA >ETZwBwZ9kcmQeB1LQ0LjigjBzQZOM7Iid8oWZMjlrXMBjAbKHafXSDFwD1iK+O9S >X0nUgI+ECtMr5Ya7k72i >=smID >-END PGP SIGNATURE- >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] KGS 8-dan and strong opponents
Dear Ingo and others, I'd like to note that Zen19A limits the handicap upto 4 stones instead of usual 6. This is the reason the account is not Zen19 but Zen19A. #It's weill known that MC bots are not good at (high) handicap games. Best, Hideki Ingo Althofer: <trinity-dc26271f-287f-4ae8-9d4a-e45ffc26e4bc-1466492847498@3capp-gmx-bs06>: >Hi, > >for the first time there is a bot with "stable" 8-dan rank on KGS: >Zen19A, an alternative Zen account. > >http://www.gokgs.com/gameArchives.jsp?user=zen19a > >Congratulations to the Zen team! > >And an observation: >Already several games against human players with 8-dan >or 9-dan rank have been played by Zen19A in the last >12 hours. > >Likely, opponents in this strong range will help the >programmers to improve their brainchild further. > >Ingo. >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Google used TPUs for AlphaGo
Some photos here. http://itpro.nikkeibp.co.jp/atcl/column/15/061500148/051900060/ (in Japanese). The performance/watt is about 13 times better, a photo in the article shows. Hideki Petr Baudis: <20160519105443.go22...@machine.or.cz>: > Hi, > > it seems that Google in fact used TPUs for AlphaGo rather than GPUs: > > >https://cloudplatform.googleblog.com/2016/05/Google-supercharges-machine-learning-tasks-with-custom-chip.html > >It's be interesting to know what the speedup factor against, say, >Tesla K40 is. In the Nature paper, they talk about GPUs, but they >could have switched to TPUs for training only after then Fan Hui matches. > > Petr Baudis >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Google used TPUs for AlphaGo
DGX-1 looks not shipped yet. TPUs has been used more than 1 year. This could be a big advantage for Google. Hideki Darren Cook: <573da3db.8000...@dcook.org>: >> It's be interesting to know what the speedup factor against, say, >> Tesla K40 is. > >Or against the P100 chip [1], which claims the same "order of magnitude" >speed-up on neural nets by doing the same thing (half-precision floating >point). > >Darren > >[1]: >http://nvidianews.nvidia.com/news/nvidia-delivers-massive-performance-leap-for-deep-learning-hpc-applications-with-nvidia-tesla-p100-accelerators > >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] KGS bot tournaments: structure
Gian-Carlo Pascutto: <5731dc19.2020...@sjeng.org>: >On 10-05-16 11:23, Hideki Kato wrote: > >> CGOS is better place for those lower programs, isn't it? > >Not really, the pool of opponents is smaller and contains no humans. It >sort of depends on what the goal of the author is. Even if she's only >interested in measuring vs other computer opponents, a KGS tournament >*may* offer a bigger pool because there's more incentive to connect at a >given time. The number of games is the most important point to get correct ratings. It takes several weeks or more on KGS to have enough (at least several handreds) games. Also, there are several hardles to get ranked on KGS. Programs have to be stable and no serious bugs, for examples. >> I'm not against creating lower division, just wonder if it's really >> necessary. Recently it's easier to implement "large patterns" which >> is necessary to beat GNU Go on 19x19 using DCNN than Remi's B-T model >> and so most programs could quickly reach GNU Go level. > >I think it's up to the author to decide which approach he or she wants >to pursue. It's not because everyone is making hand-crafted pattern >databases with elaborate rules for local tactical search, that you can't >try just playing out games randomly, for example, even if that approach >seems weak right now. Maybe it turns out to scale better in the long run. > >> If this is correct, creating two divisions might be a bad idea. > >Not necessarily disagreeing there. > >-- >GCP >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] KGS bot tournaments: structure
CGOS is better place for those lower programs, isn't it? I'm not against creating lower division, just wonder if it's really necessary. Recently it's easier to implement "large patterns" which is necessary to beat GNU Go on 19x19 using DCNN than Remi's B-T model and so most programs could quickly reach GNU Go level. If this is correct, creating two divisions might be a bad idea. Hideki Nick Wedd: <CAEVtG+M-2CzXmP=rlwg0wsm9pgq738at_cd_so9phvrez5c...@mail.gmail.com>: >At present, KGS bot tournaments are run as Swiss tournaments. Entry is >restricted to bots able to beat GNU Go, which is rated at 7k. > >Two suggestions have been made recently for changes in the structure. One >is to allow bots of any strength to enter, and to have two divisions, with >those unable to beat GNU Go in the lower division. The other is to use >McMahon instead of Swiss. > >A problem with McMahon is that the scheduling software (a module of the KGS >server, which I have no control over) uses the ratings assigned by KGS, and >many bots do not have such ratings. I don't think it's reasonable to >require all entrants to have rated bot status. Rated bot status is granted >only to *stable* bots, ones which do not change rapidly in strength and >thereby upset the ratings of their opponents. I think it's fine for >someone to enter an experimental bot of unknown strength in a tournament, >but such a bot cannot have a KGS rating. > >Maybe we could combine the two suggestions. Have two divisions: upper >division McMahon, with entrants required to have KGS ratings of 7k or >better; lower division open. I would like to hear people's views on this - >and any more suggestions they have. > >Nick -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Computer Olympiads 2016 and 2011
Dear Ingo, Yes, I will. Best, Hideki Ingo Althofer: <trinity-9dc78d90-860c-48c3-8585-78f3f3540668-1462683382851@3capp-gmx-bs78>: >Dear Hideki, > >thanks for the hint. I have updated the passage >on Jun'ichi Hashimoto. I still remember his interesting talk. > >http://althofer.de/tilburg-2011-go.html > >By the way: Will you come to Leiden this year? > >Cheeers, Ingo. >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] COmputer Olympiads 2016 and 2011
Thanks Ingo for the nice photos, The nice guy in the photo of "Go programmers" is Jun'ichi Hashimoto from JAIST (without glasses; photo with glasses http://www.gamebusiness.jp/article/2010/09/07/2094.html, although looks very differently :). He was visiting Tilburg univ for his doctor thesis (with Jaap?) in 2011 and worked harder for the events as a staff. He also had a talk on "Accelerated UCT and Its Application to Two-Player Games" in ACG2011 (http://link.springer.com/chapter/10.1007%2F978-3-642-31866-5_1, https://pdfs.semanticscholar.org/e07f/bdc8ddb7df3d03fd6e0f765815b761e9deed.pdf). Best, Hideki Ingo Althofer: <trinity-64aebc67-877b-4584-b9b1-bc52236c089b-1462634102064@3capp-gmx-bs35>: >Hello everybody, > > >(i) This year the Computer Olympiad takes place in Leiden (NL), >from June 27 to July 03. The deadline for participation >is approaching quickly: participants have to register >until May 15 (otherwise their fees double). More info here: >http://icga.leidenuniv.nl/ > >(ii) Five years ago, the ICGA Computer Olympiad took place >in Tilburg. I looked in some old folders and found a >few pictures on the computer Go part of that event. >You can watch them here: >http://althofer.de/tilburg-2011-go.html > >Ingo. >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] NHK news of UEC Cup (with video) (in Japanese)
Hi Ingo, SBS (South Korea) news. http://youtu.be/09DydFX_5OQ The venue is well seen by this. Hideki Ingo Althofer: <trinity-0eb724a6-7e27-4daa-bfb8-9d2996f4ebd2-1458565876156@3capp-gmx-bs37>: >Hi Hideki, > >thanks for the link. The video is very interesting, also for >people without any knowledge of Japanese (like me). > >Tsuchi-oto, >Ingo. > >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] NHK news of UEC Cup (with video) (in Japanese)
WoW! http://www3.nhk.or.jp/news/html/20160320/k10010450341000.html thanks. Hideki Ingo Althofer: <trinity-4c0ed4ad-bd47-4e15-8ba8-271c692afb75-1458546707764@3capp-gmx-bs68>: >Hi Hideki, > >thanks for the information. >Did you forget a/the link? > >*** >By the way: Currently there are two bots in the top-100 >of KGS: CrazyStone on rank 36 and Zen on rank 52 (both 7 dan). >http://www.gokgs.com/top100.jsp > >Ingo. > > > > >> Gesendet: Montag, 21. März 2016 um 08:41 Uhr >> Von: "Hideki Kato" <hideki_ka...@ybb.ne.jp> >> An: computer-go@computer-go.org >> Betreff: [Computer-go] NHK news of UEC Cup (with video) (in Japanese) >> >> Hi all, >> >> NHK (Japanese BBC) released the news video of UEC Cup on its web >> site. It broadcast prime time (7 pm) yesterday in Japan, I >> believe. Partly including the press conference of DeepZenGo >> project. >> #Thanks to Sunday for 2m 41s long! >> #The closing is Prof. Ito, Chairperson of the executive >> committee of UEC cup. >> #Ohashi pro (6d) is looking into the screen with me. >> >> Hideki >> -- >> Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> >> ___ >> 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
[Computer-go] NHK news of UEC Cup (with video) (in Japanese)
Hi all, NHK (Japanese BBC) released the news video of UEC Cup on its web site. It broadcast prime time (7 pm) yesterday in Japan, I believe. Partly including the press conference of DeepZenGo project. #Thanks to Sunday for 2m 41s long! #The closing is Prof. Ito, Chairperson of the executive committee of UEC cup. #Ohashi pro (6d) is looking into the screen with me. Hideki -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] UEC cup 2nd day
Dear Ingo, >Hi Hiroshi, > >thanks for the many updates. > >On another site I read that the bits on rank 1 and 2 will >play exhibition matches against a pro player on Wednesday. Yes, Koichi Kobayashi 9p. #Nick's h-c page: http://www.computer-go.info/h-c/index.html >Will those games be transmitted on KGS? I guess it's very difficult because all broadcast rights must exclusively be owned by the sponsor, "Igo Shogi Channel." >Has it been decided alreay which handicap? Basically 3 stones. If Kobayashi 9p lost the first game (vs Darkforest), the second game (vs Zen) will be played with 2 stones. #Not announced yet. Hideki >Thanks in advance, Ingo. > > >> Gesendet: Sonntag, 20. März 2016 um 07:41 Uhr >> Von: "Hiroshi Yamashita" <y...@bd.mbn.or.jp> >> An: computer-go@computer-go.org >> Betreff: Re: [Computer-go] UEC cup 2nd day >> >> Zen won against darkforest >> >> 1st Zen >> 2nd darkforest >> 3rd CrazyStone >> 4th Aya >> >> Hiroshi Yamashita >> >> ___ >> 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] UEC cup 2nd day
Attached is my _personal_ copy of the final game. All official game records will be available soon, though. #After move 295, the operator resigned. #Official results of the second day: http://jsb.cs.uec.ac.jp/~igo/eng/result2.html Hideki Pawe Morawiecki:
Re: [Computer-go] Congratulations to AlphaGo
Congratulations again, Aja! Once you told me (or I told you?) that "Go is fighting!". Now which do you think, fighting or (whole-board) perspective? Best regrads, Hideki Aja Huang: <cajbo_wgm5__ekrrk9jtx45ku-z3zrwfblmchkp5xhxegyhk...@mail.gmail.com>: >Thanks all. AlphaGo has won the match against Lee Sedol. But there are >still 2 games to play. > >Aja > >On Sat, Mar 12, 2016 at 5:49 PM, Jim O'Flaherty <jim.oflaherty...@gmail.com> >wrote: > >> It was exhilerating to witness history being made! Awesome! >> >> On Sat, Mar 12, 2016 at 2:17 AM, David Fotland <fotl...@smart-games.com> >> wrote: >> >>> Tremendous games by AlphaGo. Congratulations! >>> >>> >>> >>> *From:* Computer-go [mailto:computer-go-boun...@computer-go.org] *On >>> Behalf Of *Lukas van de Wiel >>> *Sent:* Saturday, March 12, 2016 12:14 AM >>> *To:* computer-go@computer-go.org >>> *Subject:* [Computer-go] Congratulations to AlphaGo >>> >>> >>> >>> Whoa, what a fight! Well fought, and well won! >>> >>> Lukas >>> >>> ___ >>> 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 >> > inline file >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Congratulations to AlphaGo
Congratulations to DeepMind team! What an excellent full-board perspective. Hideki Lukas van de Wiel:
Re: [Computer-go] AlphaGo won first game!
Congratulations, Aja and David! Very remarkable win! Pasky, It's too early to conclude any, I think, because no records of losing games have been published, ie., no weakpoints of AlphaGo are open. I believe that the essential problems which come from current MCTS (bottom-up) framework, such as solving complex semeai's and double-ko's, aren't solved yet. Hideki Petr Baudis: <20160309171109.gu12...@machine.or.cz>: > Hi! > >On Wed, Mar 09, 2016 at 04:43:23PM +0900, Hiroshi Yamashita wrote: >> AlphaGo won 1st game against Lee Sedol! > > Well, I have to eat my past words - of course, there are still four >games to go, but the first round does not look like a lucky win at all! > > Huge congratulations to the AlphaGo team, you have done truly amazing >work, with potential to spearhead a lot of further advances in AI in >general! It does seem to me that you must have made a lot of progress >since the Nature paper though - is that impression correct? > > Do you have some more surprising breakthroughs and techniques in store >for us, or was the progress mainly incremental, furthering the training >etc.? > > > By the way, there is a short snippet in the paper that maybe many >people overlooked (including me on the very first read!): > >> We introduce a new technique that caches all moves from the search >> tree and then plays similar moves during rollouts; a generalisation of >> the last good reply heuristic. At every step of the tree traversal, the >> most probable action is inserted into a hash table, along with the >> 3 × 3 pattern context (colour, liberty and stone counts) around both the >> previous move and the current move. At each step of the rollout, the >> pattern context is matched against the hash table; if a match is found >> then the stored move is played with high probability. > > This looks like it might overcome a lot of weaknesses re semeai etc., >enabling the coveted (by me) information flow from tree to playouts, if >you made this to work well (it's similar to my "liberty maps" attempts, >which always failed though - I tried to encode a larger context, which >maybe wasn't good idea). > > Would you say this improvement is important to AlphaGo's playing >strength (or its scaling), or merely a minor tweak? > > > Thanks, > >-- > 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Mastering the Game of Go with Deep Neural Networks and Tree Search (value network)
Detlef Schmicker: <56b385ce.4080...@physik.de>: >-BEGIN PGP SIGNED MESSAGE- >Hash: SHA1 > >Hi, > >I try to reproduce numbers from section 3: training the value network > >On the test set of kgs games the MSE is 0.37. Is it correct, that the >results are represented as +1 and -1? Looks correct. >This means, that in a typical board position you get a value of >1-sqrt(0.37) = 0.4 --> this would correspond to a win rate of 70% ?! Since all positions of all games in the dataset are used, winrate should distributes from 0% to 100%, or -1 to 1, not 1. Then, the number 70% could be wrong. MSE is 0.37 just means the average error is about 0.6, I think. Hideki >Is it really true, that a typical kgs 6d+ position is judeged with >such a high win rate (even though it it is overfitted, so the test set >number is to bad!), or do I misinterpret the MSE calculation?! > >Any help would be great, > >Detlef > >Am 27.01.2016 um 19:46 schrieb Aja Huang: >> Hi all, >> >> We are very excited to announce that our Go program, AlphaGo, has >> beaten a professional player for the first time. AlphaGo beat the >> European champion Fan Hui by 5 games to 0. We hope you enjoy our >> paper, published in Nature today. The paper and all the games can >> be found here: >> >> http://www.deepmind.com/alpha-go.html >> >> AlphaGo will be competing in a match against Lee Sedol in Seoul, >> this March, to see whether we finally have a Go program that is >> stronger than any human! >> >> Aja >> >> PS I am very busy preparing AlphaGo for the match, so apologies in >> advance if I cannot respond to all questions about AlphaGo. >> >> >> >> ___ Computer-go mailing >> list Computer-go@computer-go.org >> http://computer-go.org/mailman/listinfo/computer-go >> >-BEGIN PGP SIGNATURE- >Version: GnuPG v2.0.22 (GNU/Linux) > >iQIcBAEBAgAGBQJWs4XOAAoJEInWdHg+Znf4btkQAKP5T6o8qk9Fv1z/JJGTdcph >A8rXWXNPNybSTqBVh7IJHKB9WxNIIOODjj6JYL4vhjFC+Eqy7AcpHkFZEn+f9j2G >Kl3w6G88M9P8/iCn++pNHF3dPTrsn2doUcUjF1fJZsKeNCISuvbHwbzgXupna+lH >qeCpYx/VBQMJIdhGqTmWsQozbFMFbeTumMH94UwNkTwo4Tnue/UCJweU0bWIIt0D >TtCyLDsDcFy/qNrZC97858tpvOpo3hWs7pLf8ed+9r13UGeJhQJkedg6Oq0e5wTl >Ye36Z1/2QHnQtvUbk6yjd6GMK0lo6LOOC1lTpp1nFvzcZ4ifrY2LejQ+7nWmafRq >y46aeH4jqtEF2GXsFTq7ATftSYoeeUzgKxb8t4D8ShP3cRRWkpHFyGtdq25yYO2P >hp35zJHAlCtUUOzy0YBY+mngYnbwxjp1ykwUi5DmubtRTfhf4pTAH+DF5/UeItn/ >IebMR9JlfCsFJZ7BLq7P7UoHz0eiG2vVKNXUP0Np4LA3KAm1IAOfJWRmLP9TrHOX >32vdIQLXlbDMTfbloXrFjQo3wm3pKrKstI41Pyoo5d8kE8FPxjCga6EfmarjRKGY >UBf7Vz8iijGWjTjszui78HlJBIGls9qyuxNUcBN4+kfJJLP633HO+tmDM1hcsD4D >BPtgsamuDtf5jvxGTa9d >=Ab0j >-END PGP SIGNATURE- >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Mastering the Game of Go with Deep Neural Networks and Tree Search (value network)
I think the error is defined as the difference between the output of the value network and the average output of the simulations done by the policy network (RL) at each position. Hideki Michael Markefka:
Re: [Computer-go] What hardware to use to train the DNN
Since Zen's engine is improved sololy by Yomato, I have no idea in detail but I believe Yamato has used one Mac Pro so far (Linux and Windows). #He has implemented DCNN by himself, not using tools. Hideki David Fotland: <0a0301d15de7$1180d760$34828620$@smart-games.com>: >Detlef, Hiroshi, Hideki, and others, > >I have caffelib integrated with Many Faces so I can evaluate a DNN. Thank you >very much >Detlef for sample code to set up the input layer. Building caffe on windows >is painful. If >anyone else is doing it and gets stuck I might be able to help. > >What hardware are you using to train networks? I dont have a cuda-capable >GPU yet, so I'm >going to buy a new box. I'd like some advice. Caffe is not well supported on >Windows, so I >plan to use a Linux box for training, but continue to use Windows for testing >and >development. For competitions I could use either windows or linux. > >Thanks in advance, > >David > >> -Original Message- >> From: Computer-go [mailto:computer-go-boun...@computer-go.org] On Behalf >> Of Hiroshi Yamashita >> Sent: Monday, February 01, 2016 11:26 PM >> To: computer-go@computer-go.org >> Subject: *SPAM* Re: [Computer-go] DCNN can solve semeai? >> >> Hi Detlef, >> >> My study heavily depends on your information. Especially Oakfoam code, >> lenet.prototxt and generate_sample_data_leveldb.py was helpful. Thanks! >> >> > Quite interesting that you do not reach the prediction rate 57% from >> > the facebook paper by far too! I have the same experience with the >> >> I'm trying 12 layers 256 filters, but it is around 49.8%. >> I think 57% is maybe from KGS games. >> >> > Did you strip the games before 1800AD, as mentioned in the FB paper? I >> > did not do it and was thinking my training is not ok, but as you have >> > the same result probably this is the only difference?! >> >> I also did not use before 1800AD. And don't use hadicap games. >> Training positions are 15693570 from 76000 games. >> Test positions are 445693 from 2156 games. >> All games are shuffled in advance. Each position is randomly rotated. >> And memorizing 24000 positions, then shuffle and store to LebelDB. >> At first I did not shuffle games. Then accuracy is down each 61000 >> iteration (one epoch, 256 mini-batch). >> http://www.yss-aya.com/20160108.png >> It means DCNN understands easily the difference 1800AD games and 2015AD >> games. I was surprised DCNN's ability. And maybe 1800AD games are also >> not good for training? >> >> Regards, >> Hiroshi Yamashita >> >> - Original Message - >> From: "Detlef Schmicker" <d...@physik.de> >> To: <computer-go@computer-go.org> >> Sent: Tuesday, February 02, 2016 3:15 PM >> Subject: Re: [Computer-go] DCNN can solve semeai? >> >> > Thanks a lot for sharing this. >> > >> > Quite interesting that you do not reach the prediction rate 57% from >> > the facebook paper by far too! I have the same experience with the >> > GoGoD database. My numbers are nearly the same as yours 49% :) my net >> > is quite simelar, but I use 7,5,5,3,3, with 12 layers in total. >> > >> > Did you strip the games before 1800AD, as mentioned in the FB paper? I >> > did not do it and was thinking my training is not ok, but as you have >> > the same result probably this is the only difference?! >> > >> > Best regards, >> > >> > Detlef >> >> ___ >> 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] guess AlphaGo's CGOS rating
>But AlphaGo single machine is stronger than Fan Hui, since it won 8-2 in >all matches combined on one machine, as far as I understand they used >this version No. AlphaGo Distributed was used for the 10 games. Hideki >On 2016-02-02 13:33, Hiroshi Yamashita wrote: >> Hi, >> >> I made CGOS rating include AlphaGo. >> >> CGOSRemi's pro rating(same as paper) >> >> Ke Jie 4642? 3620 Strongest human >> Lee Sedol4538? 3516 >> AlphaGo 4162? 3140 Distributed, 1202CPUs, 176GPUs >> Fan Hui 3942? 2920 >> AlphaGo 3912? 2890 48CPUs, 8GPUs, one machine >> Zen 24core 3800??? KGS 7d, 24 cores, one machine. >> Zen-10.8-2c 3485 CrazyStone-0002 3313 i7-5930K, 6 threads >> Zen-10.8-1c 3299 DCNN versoin >> Zen-1c-2.8G 3072 >> AlphaGo RL 2780? Reinforcement learning, no search >> Aya786l_10k 2718 KGS 2d >> darkfores2 2657? KGS 3d, no search >> AlphaGo SL 2539? 1517 DCNN, no search >> darkfores1 2490? KGS 2d, no search >> pachi11_Pat_100k 2478 >> darkforest 2271? KGS 1d, no search >> DCNN_Detlef542179 Detlef's 54%, no search >> pachi11_Pat_10k 2104 >> DCNN-Detlef 1903 Detlef's 44%, no search >> Gnugo-3.7.10-a1 1800 KGS 5k >> >> ? is guess. ??? is my anticipation. >> >> This result is based on AlphaGo(RL) beats Pachi 100k with 85% winrate. >> darkforest is also based on winrate against Pachi 100k. >> Pro and computers comparison is maybe only from 8-2 result AlphaGo vs >> Fan Hui. So pro is maybe more strong (or weak). >> Watching this table, Zen 7d maybe play nice game against one machine >> AlphaGo. >> >> Pro rating is from Remi's site. >> http://www.goratings.org/ >> CGOS rating is from BayesElo. >> http://www.yss-aya.com/cgos/19x19/bayes.html >> >> Regards, >> Hiroshi Yamashita >> >> ___ >> 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 -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Mastering the Game of Go with Deep Neural Networks and Tree Search
Ingo Althofer: <trinity-a297d40e-3cf2-45f1-8d38-13a5912b636c-1454339862588@3capp-gmx-bs72>: >Hi Hideki, > >first of all congrats to the nice performance of Zen over the weekend! > >> Ingo and all, >> Why you care AlphaGo and DCNN so much? > >I can speak only for myself. DCNNs may be not only applied to >achieve better playing strength. One may use them to create >playing styles, or bots for go variants. > >One of my favorites is robot frisbee go. >http://www.althofer.de/robot-play/frisbee-robot-go.jpg >Perhaps one can teach robots with DCNN to throw the disks better. > >And my expectation is: During 2016 we will see many more fantastic >applications of DCNN, not only in Go. (Olivier had made a similar >remark already.) Agree but one criticism. If such great DCNN applications all need huge machine power like AlphaGo (upon execution, not training), then the technology is hard to apply to many areas, autos and robots, for examples. Are DCNN chips the only way to reduce computational cost? I don't forecast other possibilities. Much more economical methods should be developed anyway. #Our brain consumes less than 100 watt. Hideki >Ingo. > >PS. Dietmar Wolz, my partner in space trajectory design, just told me >that in his company they started woth deep learning... >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Mastering the Game of Go with Deep Neural Networks and Tree Search
Olivier Teytaud:
Re: [Computer-go] Zen19X achieved stable KGS 7d
Aja Huang:
Re: [Computer-go] Zen19X achieved stable KGS 7d
Thanks, it's very reasonable :) Hideki Michael Markefka: <cajg7paoabgu6h6swp-180us9qtcovffgz3tqfsm3sashqq_...@mail.gmail.com>: >On Mon, Feb 1, 2016 at 1:44 PM, Hideki Kato <hideki_ka...@ybb.ne.jp> wrote: >> I was, btw, really surprised when Zen beat fj with two stones >> handi. >> http://files.gokgs.com/games/2016/1/31/Zen19X-fj.sgf >> >> Hideki > >On the DGoB forums fj stated, possibly in jest, that this was an even >game, as he had had a glass of wine for every stone of handicap there. >Seems like he wasn't in the most competitive of moods. Still a good >game though. :) >___ >Computer-go mailing list >Computer-go@computer-go.org >http://computer-go.org/mailman/listinfo/computer-go -- Hideki Kato <mailto:hideki_ka...@ybb.ne.jp> ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go