Japanese title matches without Iyama

[Uberdude]

Shibano won the 3rd game by resign.

[Uberdude]

Shibano is currently playing Kyo Kagen in the 67th Oza challenger decision match (he beat Cho U in an earlier round). Can he have a go at another title or will Kyo get a chance to regain one after losing the Gosei to Hane Naoki. Iyama Yuta is the defending champion.

https://home.yikeweiqi.com/#/live/board/22976

(I wasn't sure whether to put this here or the old Shibano thread, but it seemed topical ).

UPDATE: Shibano won by resign. He was behind for much of the game but in the last middlegame area after living with his weak groups he managed to reduce Kyo's moyo more than the bots thought he deserved.

[Uberdude]

Shibano seems to be getting the hang of two-day title matches: he just won the 4th Meijin game to take a 3-1 lead so Cho U faces kadoban. Next game is 7th and 8th October.

[Uberdude]

Meijin 5th game live now. If Shibano wins he is the new Meijin, would it be youngest ever? When Iyama beat Cho 10 years ago to win his first big title he was 20 years 4 months. Shibano turns 20 in November.

https://home.yikeweiqi.com/#/live/room/23772/1/22107875

[Uberdude]

Shibano won and is thus the new Meijin at one day shy of 19 years and 11 months.

[bernds]

Shibano won and is thus the new Meijin at one day shy of 19 years and 11 months.

A wise man once said "A Meijin in his twenties is inconceivable".
Although I guess Shibano will get there in a month.

[sorin]

Shibano won and is thus the new Meijin at one day shy of 19 years and 11 months.

A wise man once said "A Meijin in his twenties is inconceivable".
Although I guess Shibano will get there in a month.

Shibano's win is just another reminder that go is not an art: it is a sport for young people

[jlt]

Move 224 illustrates that AI with too few playouts shouldn't be trusted. After a few hundred playouts, LZ157 thinks that Black leads by 80%, but after more thinking, LZ is completely confused and thinks that almost every point has about 9% winrate.

Capture.PNG (472.25 KiB) Viewed 3854 times

[Uberdude]

Move 224 illustrates that AI with too few playouts shouldn't be trusted. After a few hundred playouts, LZ157 thinks that Black leads by 80%,...

Yes, evaluations on low playouts are less reliable, particularly in positions which are not quiet, positional judgement type ones, but have long sequences which need reading out and who is winning changes sharply based on what happens many moves down the line. Ko fights, ladders and semeais being such positions bots need extra playouts to judge well.

but after more thinking, LZ is completely confused and thinks that almost every point has about 9% winrate.

Doesn't look confused to me: looks like it now correctly understands black is hopeless and every move is bad.

[Uberdude]

Here's the game with KataGo analysis (but Lizzie 0.7 puts the output in a new LZ sgf tag instead of comment so eidogo player doesn't show it): it had trouble understanding the ko at the end (thinking black was winning): failing to see that white had enough threats if he played point-losing ones which allowed him to win it.

[ez4u]

Here is Uberdude's file with the LZ tag replaced with C.