Chess Soul: Buy the Shijia chessboard at the beginning

Kuwahara 9th Dan kept mumbling in astonishment, "That shouldn't be the case, it doesn't conform to the rules of chess."

Yu Liang, who always thought he knew Tian Minze very well, fell silent, and he couldn't explain why.

Because from the situation, Tian Min seemed to be giving up, throwing chess pieces randomly on the chessboard.

On the contrary, AlphaGo did have a very structured start, with Xing Xiaomu and Xiao Fei guarding the corner, which are still common routines in AI Go.

Tian Min got up and walked to the window to get a glass of water and calm down.

Yu Liang's introduction was very easy to understand, and even the Internet tycoons in the audience who didn't know much about Go started talking about it.

On the contrary, the chess fans and all the staff in the viewing room were already excited, because as long as the black chess piece was played, three large blocks of chess pieces surrounding the target had been completed on the chessboard.

Tian Minze is not lagging behind on the ground. He wants to counter the limited number of chess pieces that attract Alpha Dog's attention. He also chooses the kind of placement that is difficult for the opponent to capture at once.

Slowly connect a series of chess pieces on the outer circle of the chessboard, trying to surround this string of chess pieces into a large "encirclement". At the same time, place some scattered chess pieces in other corners of the chessboard to distract the alpha dog.

As the black piece fell on the 207th move, the AlphaGo team in the background saw that the program had entered into a long period of thinking. After 20 minutes, the winning rate dropped directly from 90% to 40%.

Then the computer started to suffer from old habits and made a move that he thought was a general.

Kuwahara 9th Dan was naturally not that keen. He just felt that the whole chess game was strange, and he couldn't explain why it was strange. "What kind of bug?"

Tian Minze's method of play goes beyond the scope of Go. The purpose of this kind of play is too obvious, which is to enclose a large area with chess pieces.

"Did Tian Minze find AlphaGo's BUG?" Yu Liang guessed.

There was no way to continue the on-site commentary. After all, this was beyond the scope of Go. However, the Chinese Chess Academy was also prepared and soon invited an AI engineer from Tomorrow Information.

As the engineer expressed his judgment and explanation, everyone gradually understood that the AI ​​had never seen this kind of trick before, so it was naturally unable to fight against it.

In short, Tian Minze’s strategy is as follows:

In the preparation room backstage, everyone in AlphaGo's team was also a little unbelievable. The winning rate given by the AI ​​​​has reached 90% for their side, but it is obvious that the black pieces on the chessboard are creating more and more holes.

For human chess players, as long as the opponent masters the basic rules of Go, there is basically no chance of winning using this method. Because as long as your "encirclement circle" has not been closed, the opponent can easily break the entire encirclement within a few moves. At this point, because you have wasted too many pieces on the encirclement, it is no longer possible to win the game.

But for the AI, the chess pieces played by Tian Minze were not very efficient, so the AI ​​paid no attention to it and was still looking for a position with higher piece efficiency to play.

However, Gao Yu's fourth-dan position still expressed his opinion, "As an average chess player, I feel that Tian Minze took too many chances today."

Yu Liang introduced, "The winning rate of Tian Minze's chess is very low, so Alpha Dog plays the position with the highest winning rate, thus ignoring Tian Minze's small moves on the side and in the corner. This method is used to deal with humans. The chess player is useless, but it seems the AI ​​fell into the trap."

It has to be a professional commentator who can speak.

Two six zero three: c zero two four: c zero zero a: c four one one:: one one two zero

Alpha Dog has been struggling with the pros and cons of some chess shapes, but he didn't seem to notice that Tian Min had quietly surrounded a large area.

Although Tian Minze adopted an unconventional way of playing chess, it could be seen that he was still very careful and hesitated every step. On the contrary, the AI ​​​​played very quickly, almost without thinking, as if the chess game in front of it had no difficulty for it.

But Dr. Huang, who was sitting in front of the chessboard, looked extremely ugly. He had clearly felt that a bug in the computer program had been discovered.

Alpha Dog turned a blind eye to Black's moves and was unable to identify any potential dangers. He did not destroy Black's "encirclement" before it closed and allowed Black to play its part.

Without the ability to judge as a basis, the AlphaGo program can no longer run. "

Tian Min stood up again, shook hands with the shocked Dr. Huang, and walked out of the game room.

Most of what everyone had prepared before was a press release after Tian Minze's loss. Suddenly, it was replaced by a press release about Tian Minze's victory, which caught everyone off guard.

Tian Min then walked into the post-game interview booth, and it was still a routine interview at the end of every game. But this time, I saw all the reporters hurriedly deleting their articles.

There were media outlets of all sizes along the way, including news channels, sports channels, and entertainment channels. Various reporters were sandwiched on both sides of the road. When they saw Tian Minze passing by, some people started to applaud involuntarily.

Although no one could understand Tian Minze's Go game, the commentator still did not dare to comment easily. After all, the person sitting there was the number one player in the world today, and many people could not understand his Go game.

But Tian Min ignored it at all and continued to finish. After another 10 moves, AlphaGo calculated that its winning rate had dropped to 0%. After five seconds, in the 223rd move, White surrendered. burden.

Facing the camera, the young engineer was still a little nervous, but he soon became eloquent when talking about his professionalism, "Because AlphaGo was developed based on adversarial exercises. In its adversarial exercises, it has not experienced Having played enough of these moves, he lacks samples.

Soon there was thunder in the palms, and everyone had some excitement and admiration.

Because AI seems to have failed to keep up with Tian Minze's pace.

For the first time in the human-machine war, it ended with a human victory. Although this victory was a bit sudden and a bit opportunistic, it was a victory after all, and it was a key victory that human chess players needed.

Behind this unstructured approach lies some of Tian Minze's logic. It is obvious that he often takes the lead.

After saying this, several people seemed to understand.

Gao Yu sighed with emotion, "Tian Minze used magic to defeat magic."

But to say that he took the lead is an exaggeration. This is obviously whether he went to the corner to play a move, which is hard to understand. However, from the situation, it seems difficult to say that Tian Minze has no chance at all.

Sang Yuan and Yu Liang laughed at the same time.

Tian Minze didn't have much joy, and remained calm as usual.

The reporter's first question was, "Tian Jiuduan, have you found Alpha Dog's Achilles' heel? Will you consider using the same method in the next game?"