Since the emergence of "Afagou", humans have been talking about the power of Go AI. And because Go AI's leading mankind seems to be almost all-round, especially some of the characteristics of some machines, such as never tiredness, never making mistakes, etc., this is simply a weakness that humans cannot overcome.

It is precisely for this reason that many people have begun to feel desperate about Go, believing that with the update of Go AI software and hardware, the gap between humans and machines will only increase.

So the question is, is Go AI really invincible? Does humans really have no chance when facing Go AI?

Li Xiangping personally believes that the answer to the first question should be very clear, and of course Go AI is not invincible!

Because the power of Go AI is only compared to humans. In other words, the emergence of Go AI only contrasts the weakness of humans, and contrasts that before Go AI emerged, humans' understanding of Go was actually very pale and powerless. However, in front of the real God of Go-----

If there is really a god of Go in this world, then Li Xiangping believes that Go AI should be equally weak, and even the gap between Go AI and Go God should be about the difference between humans and Go AI.

It is precisely for this reason that Li Xiangping is actually a negative attitude towards the second question. He believes that humans may not have no chance when facing machines.

Of course, the "opportunity" mentioned by Li Xiangping does not mean that humans may beat machines and become crushing Go AI, but that if humans can really improve their understanding of Go after learning, and in a specific situation, for example, when a human chess player is fighting against Go AI, the human chess player's character explodes. He overcomes the unique physiological and psychological problems of human chess player and uses this to offset the physical advantages of the machine.

So in this case, Li Xiangping believes that human chess players should still have a chance to win.

But another question is, how should humans learn when facing Go AI?

First of all, humans cannot learn whether machines do not make mistakes. Secondly, "Monte Carlo" and "Deep Learning" of Go AI, of course, these things cannot be learned.

In addition to these, there are also people who say that Go AI's judgment is better than humans and can learn their judgment ability. However, if you think about it carefully, you will find that this is simply a false proposition. The "judgment" of machines. Can this thing be learned by humans?

Obviously, the judgment of machines is also unlearnable.

Therefore, after thinking about it, you will find that the only thing humans can learn from machines is actually the "dog tricks" they have made. They carefully think about the "dog tricks", and then analyze and compare the "dog tricks" with the humans' own methods, clarify the connotation of the "dog tricks", find out the gap between the two, and integrate the connotation of the "dog tricks" with the humans' own thoughts.

Li Xiangping believes that if this can be done, humans may have a chance to defeat Go AI.

This is a little insight that Li Xiangping learned during the preparations for the previous month.

This is actually the reason why he used this "Little Megu Erjian's high jump angle" method today, and then he was able to make a skillful move when he was more than 70 moves and gained the lead in the situation.

Because through previous in-depth research, Li Xiangping found that although it was so simple and two-handed chess, one that looked nothing special, compared with the usual playing methods used by humans, he would find that the concept of Go AI is indeed very different from human thoughts----

It must be mentioned in particular that Go AI actually has no "ideology", it only has algorithms, but Li Xiangping believes that this does not prevent humans from "translating" machine algorithms into human concepts, and then comparing the two.

Take this "Little Mean Erjiao's high jump angle" as an example. If this is considered a "dog move", then the recent method of human beings and it is the classic "Worry-free corner", and the other is "Little Mean Big Flying Corner". If you compare these three, you will find some interesting things.

Let’s first talk about the most classic modern human Go, which is actually the most familiar "Worry-free Cat".

It is no exaggeration to say that since Go went east and the Japanese abolished the throne and started playing Xiaomu Go, "Wuyoujiao" has been the core of the entire "Xiaomu Go" system.

When two chess pieces are swung to the corner, they can surround the real space with a height of over 10 eyes in a solid and safe and safe manner. This is human beings’ understanding of this chess shape, and this is actually the origin of the name “Worry-free horn”.

Why did this method last forever in the human Go field? From the beginning of the Japanese Taoist sect to the beginning of Afago, why did "Wuyoujiao" become popular in the human Go field for hundreds of years?

Please pay attention to the keyword "beyond 10-eye". The two-handed chess is "beyond 10-eye", so is the one-handed chess about 5 and a half eyes?

Speaking of this, I believe many people suddenly realized it? In the early years, modern Go was popular for a long time.

The so-called rule of posting is actually a problem of human understanding of the effectiveness of Go pieces. Humans have been walking for hundreds of years from no posting to black posting 5 and a half eyes, so why do they stay for so long on black posting 5 and a half eyes? In fact, in humans' understanding, the so-called first-hand efficiency, the value of a chess move is about 5 and a half eyes, so since you are holding black first, you have to deduct your 5 and a half eyes to compensate the other party.

And this is actually the reason why "Wuwujiao" is so popular.

Since I can play "over 10 points", it proves that my two pieces are effective, not to mention that they are particularly efficient, but not to seek merit but to seek nothing wrong. At least these two pieces are completely sufficient, which should be considered very correct and no one can blame.

Well, since the popularity of "Worry-free Corners" is related to the rules of posting, it is naturally easy to understand if you use the same idea to consider the method of "small eyes, big flying corners" to consider.

This method of playing is mainly popular in the era of large-scale maps. Since the end of the last century, humans began to artificially modify the rules of Go, and black maps with 5 and a half eyes have turned into 6 and a half eyes or even 7 and a half eyes. In this case, everyone found that the effect of "Worry-free Corner" seems to be a bit insufficient, so what should I do if it is not enough? It is very simple. I will surround the whole way, and maybe I can surround the other way, which will just meet the new surrogate standards.

So in this way, the method of "small eyes fly into big corners" came into being.

After talking about these two common methods of human use, let’s talk about the “Little Meguji High Jumping Angle” that Go AI likes to play.

There is one difference that is worth noting. Among the two human methods just now, the thing that everyone uses to measure subeffects is relatively simple, that is, the real-world and the number of mesh.

However, what about this "small eye two high jump angle"? It seems that it cannot be measured by a simple number of mesh. It is said that it has no image. It is obviously both pieces on the corner, but it is necessary to say that it has a certain degree of realism from the beginning. Both pieces are at a high position and seem to be leaking everywhere. It can be said that as long as you have more than 3 amateur chess power, there are N methods to break such chess shapes.

But it is precisely because of this reason that such a chess shape can only be a "interest" in human Go, but not a "normal" because in human Go thinking, we actually hate this ambiguous and vague thing. And in human Go theory, we actually reject such method of playing. This method of playing is called "half-way, specious but not true."

For example, Li Xiangping has seen a lot of chess comments written by Mr. Wu Qingyuan and Mr. Hideyuki Fujisawa. These two human chess players have a lot of chess principles, which are very repulsive to this method of placing grounds that is not like placing grounds, and taking momentums is not like placing momentums. Although these two people face the same situation, the two of them have different methods of placing themselves. Fujisawa's old charlatan prefers placing heavy taste, while Mr. Wu's choices tend to be extreme, either extreme grounds or extreme momentum, but they are a little bit apart from both, but neither of them are all very repulsive.

When encountering such a chess, the old Fujisawa charlatan would say that he had "stomach pain", while Mr. Wu would say "half-way" and "half-way".

Not only that, but also very interesting things. In fact, for such chess, human rejection is not only Japanese and modern chess, but also medieval chess.

For example, the master of chess in the middle age, namely Li Xiangping's current plug-in Shi Da Chess Saint, said in his "Summary of All Meetings" that year: Everything should be fought for nothing.

Needless to say, if we use this standard to look at the chess shape of "Xiaomu Erjian High Jumping Settings", it obviously does not meet the chess principles requirements of the medieval chess.

Which is better, human chess theory or Go AI? To be honest, Li Xiangping doesn’t really understand it now.

However, there is no way, who can make Go AI higher? Therefore, even if Li Xiangping doesn't understand it at all now, he can only lean towards Go AI.

If Li Xiangping had to say good things about this method, he could only say this: human methods tend to be more static, while Go AI methods may be a more advanced dynamic method that makes the chessboard full of more possibilities.

Of course, Li Xiangping still cannot master this dynamic approach, but it does not prevent him from imitating him now. He is working hard in this direction now.

Luckily today, Li Xiangping successfully imitated the whole situation for the first time. Since more than 70 men took the "last" skill, he has been firmly controlling the whole situation.

There was no big wave behind this game. At around 5:00 pm, Li Xiangping won the first victory with 3.50 goals.

Finally, Li Xiangping breathed a sigh of relief.
Chapter completed!