Re: [Computer-go] Zen beat Ha, Yongil 5p with 4 stones (Re: TAAI details?)
Hello Hideki, thank you for the explanation and the info. Stefan Kaitschick: The silly p3 wasnt really a problem, imo that was just a bot-typical let's go home move. P3 was played due to a hardware error (perhaps floating point unit or memory), I strongly believe by observing the log file. ... Time for handi 3 ... Many people suggest (or encourage :) to move 3 handi, however, Zen lost all six games against three professonals on 9x9 board today. I think that small boards (particularly 9x9) are another world than 19x19. And there is a saying (at least in chess): You learn most by losses. The game records can be downloaded from http://entcog.c.ooco.jp/entcog/event/kifu_20121125.lzh . I can not read that format. Can someone please translate it into sgf? This result shows the potential power of professional players when play seriously. Were these humans payed according to their results? Ingo. ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] Zen beat Ha, Yongil 5p with 4 stones (Re: TAAI details?)
As Hiroshi wrote, the name of the players are swapped in the first game, sorry. Hideki Erik van der Werf: CAKkgGrMakHiZuUsY9PPgHsX=bkck1-n3nlrjoriqlyduwws...@mail.gmail.com: On Sun, Nov 25, 2012 at 2:18 PM, Hideki Kato hideki_ka...@ybb.ne.jp wrote: Zen lost all six games against three professonals on 9x9 board today. Hi Hideki, Didn't it win the first game? Erik ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go -- Hideki Kato mailto:hideki_ka...@ybb.ne.jp ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] Zen beat Ha, Yongil 5p with 4 stones (Re: TAAI details?)
Thank you Hiroshi. The names of the players are swapped, sorry all. The games were: 1) Ichiriki 2p (B) vs Zen, 2) Ohashi 5p (W) vs Zen, 3) So 8p (B) vs Zen, 4) Ichiriki 2p (W) vs Zen, 5) Ohashi 5p (B) vs Zen, 6) So 8p (W) vs Zen. The time setting was: 20 min main time and 30 second for every move (no extra byo-yomi). Chinese rules and 7.0 komi were used. Zen's hardware was 4 pc cluster (12, 6, 6 and 6 cores) at 4 GHz, same as TAAI and Zen19D/S on KGS. Professinals used much longer time than Zen; more than 20 min in four games of six while Zen used 10 or 15 min for all games. Sometime they thought more than 5 min (upto 10) for a move, move 11 of the first game for example. On the first game, Zen would play J4 (hane) next at move 16 (cut) but actually played G3 and went into Sekito-shibori (two stone edge squeeze) course. Either showed about 80% winrate but Ichiriki 2p said J4 could lead a safer win for W. Actually Zen had sure chances to win; hard luck. Zen had chances to draw in some games in Black but selected much risky (actually losing) moves. I guess this caused by implementing draws by adding a third value, 0.5, to UCB. To play a draw move, all better looking moves (by prior) have to be refused, or proved worse than 0.5. This could take pretty long time in some positions. #Faking komi to 6.5 might help but a better solution possible? Hideki Hiroshi Yamashita: 313F02D8CF2F41D6AD89D1A42486DD59@x60: Hi, The game records can be downloaded from http://entcog.c.ooco.jp/entcog/event/kifu_20121125.lzh . I can not read that format. Can someone please translate it into I changed it to zip. http://www.yss-aya.com/kifu20121125.zip I was impressed So Yokoku 8p comment, Zen is apt to make a mistake when we make two places that may become ko. It makes game complex. Actual playing ko makes game simple. Didn't it win the first game? Sgf's player color is wrong. Black is Ichiriki 2p in first game. Zen lost all 6 games. Regards, Hiroshi Yamashita ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go -- Hideki Kato mailto:hideki_ka...@ybb.ne.jp ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] Zen beat Ha, Yongil 5p with 4 stones (Re: TAAI details?)
On Nov 25, 2012, at 5:31 PM, Hideki Kato hideki_ka...@ybb.ne.jp wrote: Zen had chances to draw in some games in Black but selected much risky (actually losing) moves. I guess this caused by implementing draws by adding a third value, 0.5, to UCB. To play a draw move, all better looking moves (by prior) have to be refused, or proved worse than 0.5. This could take pretty long time in some positions. #Faking komi to 6.5 might help but a better solution possible? I think there is naturally two objectives the bot can go for: • Maximize chances of winning • Maximize chances of not losing I can think of a few candidate ways to handle the dual goals. To simplify the bullets below, I'll use an input win rate of W. The most logical choice would be the win rate of the last move played in the game, but there may be less noisy alternatives. 1. Use both metrics in separate threads 2. if W x, metric = P(win) else metric = P(not lose) 3. Weighted sum. metric = f(W)*P(win) + (1-f(W))*P(not lose). I would compare f(W)=W and f(W)=0.5, but more complex options may work better than both of those. 4. If all moves have P(not lose) x, metric = P(not lose). Otherwise, restrict moves to P(not lose) = x, then select based on metric=P(win) Items 1-3 can be applied to tree search. Items 2-4 can be used for final move selection. ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] Zen beat Ha, Yongil 5p with 4 stones (Re: TAAI details?)
Hi Hideki, thanks for the information, and congratulations to the new success of Zen. My impression is that Zen (but not programs in general) has no problems to beat human professionals with 4 handicap stones. So, time seems to be ripe to have a try at handicap 3. Of course, as Stefan Kaitschick already pointed out, this (the first try at handicap 3) should be done in a nice and politically correct way. Ingo. ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] Zen beat Ha, Yongil 5p with 4 stones (Re: TAAI details?)
Very nice game for Zen. Move 40 almost won the game just by itself. White destroyed b territory so quickly, that Zen decided to show some of its power against whites weak stones. The silly p3 wasnt really a problem, imo that was just a bot-typical let's go home move. Time for Handi 3 games. Stefan ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] Zen beat Ha, Yongil 5p with 4 stones (Re: TAAI details?)
Hideki Kato: 50ad8eb2.4264%hideki_ka...@ybb.ne.jp: Also, there will be a professional (Ha, Yongil 5p in Kansai Ki-in; 4 handicap) vs Zen game at Nov 23rd (a holyday for workers) on KGS, as an event at Neyagawa Shogi and Igo Festval. I don't know the detailed schedule but the reharsal will start at 1 PM JST (UTC+0900) on KGS. http://www.igoshogineyagawa.org/ (in J) Ha 5p: http://kansaikiin.jp/profile/cgi/profile.cgi?id=hayonniru (in J) Zen won the event game against Ha, Yongil 5p by resignation. Settings were: Japanese rules, 4 handicap stones, 0.5 pts komi, 30 seconds for a move with no main time (time was counted locally due to the lag by an agent for Ha pro). The hardware for Zen was the same as for TAAI (a 4-pc cluster, all run at 4 GHz). The link to the game record is http://files.gokgs.com/games/2012/11/23/ZensGuest-Zen19S-5.sgf . I think Zen's play starts to look really impressive. It was hard to detect the usual relative weaknesses of MCTS bots. Maybe Ha Yongil did not put open enough simultaneous battlefields. But the ruthless play in top left, when tsumego is not the current bots' specialty, or its way of taking profit while harassing the invading group, are good testimonies of the bots' evolution on the two latter years I think. Jonas ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
