"What? Go? Third brother, aren't you?" Xiao Nie cried out, "The modeling of Go is more difficult than that of Chinese chess, and the modeling of chess is not a little more difficult!"
Although Xiao Nie is inferior to Yu Xing in terms of technique, awareness, and hand speed, his vision is absolutely fine. It is indeed true to say that Go modeling is difficult.
This problem is not only understood by Xiao Nie, but also by Ni Ni, who is the least skilled among the three. Remember that when she was in junior high school, she vaguely heard about the fact that the most talented people in the computer industry always wanted to conquer Go modeling.
But it never happened.
"Classmate Jin, are you sure two days is really enough?" Ni Ni asked in a low voice, "I'm afraid it won't work, right?" In view of Yuxing's excellent performance in the past, she actually saved face by saying this.
However, when asked by Ni Ni, Xiao Nie calmed down before Yuxing could answer. Thinking that Yuxing never fought without confidence in the past, he asked as if to strengthen his confidence: "Third brother, what are you doing?"
Are you sure it can work?”
Yuxing waved his hand and said, "Don't worry, I've already thought about the general direction."
Xiao Nie suddenly became excited and said: "That's alright, let's do it!" I have to say, his words really sounded like a newborn calf not afraid of tigers.
Indeed, whether it is modeling or programming, the difficulty of creating a Go game capable of human-computer dialogue is far more difficult than that of other chess games.
This point actually starts with the rules of chess. Every chess game has its own fixed rules. For example, the horse's feet in Chinese chess, and the queen in chess can move horizontally, vertically, and diagonally. These are the movements of chess pieces in specific chess games.
Basic rules. These things must be taken into consideration when designing a program.
But what are the rules for Go pieces? That is, there are no rules. As long as there are 361 spots on the Go board, they can theoretically be clicked in. Of course, eye positions do not count. But in this way
Then the problem arises. There are so many choices for where to place each chess piece, far more than in chess.
Take Chinese chess as an example. The rook moves in a straight line and the horse crosses the sun. If the computer itself completes the judgment, it can only choose a few points. Naturally, its rule model will be relatively simple.
Therefore, people can create "Deep Blue" to deal with chess masters. Human energy is limited, and brain power is also limited, so it is possible to make mistakes when consuming brain cells. But computers are different, they are tireless, as long as they follow the instructions
Just calculate according to the rules of the game.
A computer with Deep Blue, when playing chess with others, can calculate all the changes in the twelve moves made by both the enemy and ourselves, and list the gains and losses of each change in the form of scores.
When there is a pair, it will automatically play the one with the highest score, which is the most beneficial change to the chess game.
In the face of such a system, unless humans have more powerful computing power than computers and are guaranteed to make no mistakes, how can they be opponents of computers? Even if the hardware performance is improved, it can calculate more changes in the future.
But for people, it is rare to see all the changes within ten moves when playing chess.
However, Go is different. Its game rules are more humane. A master who has entered Duan is not only super capable of calculating chess moves, but also his judgment of "potential" is far better than that of a computer.
The so-called "potential" refers to the overall view. People with biological neurons may easily understand this overall view, but computers cannot. The most basic things in computers are still 0 and 1. It can only pass through the sum and
To determine whether an action is right or wrong, there is no such vague concept as the overall situation.
Therefore, for a long time, IBM has built a super chess computer "Deep Blue", but it has been unable to build a super Go computer. The reason is very simple. If the Go system is still built according to Deep Blue's digital model, then the entire
Supercomputer hardware will be overwhelmed.
After all, moving a piece is a matter of choice and judgment. As far as chess is concerned, the number of points that a chess piece can fall in each move cannot exceed 63. Excluding some points that are simply unreachable under the circumstances, the average number of points in each move is
There are only about sixteen points to choose from, and this kind of selection calculation is still within the range that a supercomputer can bear.
But Go is different. From the beginning to the end, the average number of possible positions for each move is around two hundred. With so many options, if calculated based on the computing power of Deep Blue Computer, the time it takes for it to move a move is
It takes about a year and a half, and this move is only a generally correct move. In other words, any amateur Go master can play the game so that the computer cannot find the North.
The most fundamental reason for such a situation is that computers simply do not know how to judge the general trend. Simply put, they do not know how to dig holes in advance to harm people. Therefore, computers will not be able to gain the upper hand over humans in Go at least for a long time.
Now that Yuxing wants to break this routine, it is by no means an easy thing to do. After all, the computer provided by the organizer is not a supercomputer, but a teaching computer that is slightly better than a home computer. Therefore, once the system is built, it can play Go with others.
, if you want to win by pure calculation, you basically don’t even think about it.
Therefore, when the hardware environment is really not that good, Yuxing must make the software system have a rudimentary overall view. Simply put, it is to make the program have the lowest level of intelligence. In this way, plus the computer itself is countless times more powerful than humans.
In terms of computing power, Yuxing estimates that this system can compete with amateur chess players no matter what!
The trio of Yuxing just started working, and after muttering for a while, they divided up the work. Yuxing was responsible for the modeling assumptions of graphic judgment and fuzzy judgment, and Xiao Nie was responsible for the modeling assumptions of the basic Go rules.
, as for Ni Ni, she is responsible for the modeling assumptions of life and death judgment.
It has to be said that fuzzy judgment is the most difficult part of the modeling hypothesis, because its core is to make the computer intelligent and have an overall view.
During the entire process of modeling assumptions and composition, Yuxing never touched this part of the assumptions and design. In the end, he only wrote two words there - top secret!
After Xiao Nie, who was responsible for modeling the basic Go rules, figured out his work, he stretched his neck and came over to take a look, and immediately shouted: "Third brother, top secret, what do you mean?"
"That's the literal meaning!" Yu Xing curled his lips and said, "It's a secret that must never be told, at least not to let the foreign devils know."
Xiao Nie thought about it carefully and said: "That's right. Even if the foreign devils' jury threatens not to score the system we made, this part of the design and code must not be disclosed to them!"
"I agree too!" Ni Ni beside her interjected: "But when you return to China, you can't hide your secrets. You have to give us this part of the design for reference."
"Yes, Xiaonizi is right!" Xiao Nie agreed.
Yuxing couldn't laugh or cry, and had no words to agree, but secretly thought in his heart: You want to learn, and I still want to learn! In fact, for this part, he planned to ask Yinsha to send a preliminary copy of the intelligent program to pick it up.
Beside, Ni Ni glared at Xiao Nie and asked, "Who do you think is Xiao Nizi?"
Xiao Nie deliberately didn't look at her and said sternly: "Whoever agrees to what I say!"
"you……"
Ni Ni was choked so much that she wanted to reach out and pinch him in anger. Xiao Nie was startled and quickly got out of the way.
Seeing this, Yuxing hurriedly greeted him: "Okay, okay, hurry up and do your work, the judges are here."
The two stopped talking for a moment and looked towards the rostrum. They were shocked to find two old guys patrolling among the crowds of people in each modeling group, pointing from time to time.
Yuxing's trio was almost at the back of the venue, near the corner. The two old foreigners only looked at each other at the front and then turned back, not caring about Yuxing's modeling situation.
"Third brother, we have been underestimated!" Xiao Nie said unhappily.
"That's the best!" Yu Xing sneered, "Do you know why Zhuge Liang always wins battles in the early stage and suffers in the later stage?"
Xiao Niqi said: "Is there still such a statement? Tell me, how can this happen?"
Ni Ni was also very interested. She pricked up her ears and waited to hear Yuxing's explanation.
Old God Yuxing Zaizai said: "From the novel, Kong Ming was a military god in the early days after he came out of the mountain, and he was invincible! But since he came out of Qishan to attack Wei, except for Jiang Wei, he has made almost no achievements. Instead, he has reduced the national power of Shu.
It’s exhausted…”
"Is there any explanation for this?" Ni Ni frowned.
"Of course there is a theory!" Yu Xing said, "Those people Zhuge Kongming defeated were defeated in the first frontal battle, and then suffered a complete defeat, such as Zhou Yu and Meng Huo. Looking at the Wei State, after the defeat of Chibi, the Wei army was already familiar with it.
Kong Ming’s tactics have never suffered major failures again.”
Xiao Nie's eyes lit up and he said: "Third brother, what do you mean is that Zhuge Liang was underestimated by his enemies in the early stage?"
"Yes, just like us, those foreign devils don't know how powerful we are, let them take a good look at it when the time comes!" Yu Xing glanced sideways at the foreign devil judges on the stage, his eyes twinkling.
Only then did Ni Ni understand that if she dared to fall in love with Yuxing, she was waiting here!
"Let's get to work!" Yuxing greeted the two of them and continued to immerse himself in the design.
Xiao Ni Ni was naturally full of energy and racked her brains to work hard.
At noon, the three of them hurriedly went out to have a meal, and then came back to continue their work. At four o'clock in the afternoon, they had completed the inspection of the entire model. Of course, the top-secret part was still top-secret, and they were just inspecting it.
It was directly stamped with the word "passed".
Before Yuxing and others completed the model inspection, seven or eight groups had handed over their modeling project outlines to the judges, received the keys to the small computer room, and left early to complete the programming phase.
The three Yuxings also submitted their outlines. After the judges saw the questions, they called Yuxing, who had ranked among the top ten in the previous five rounds of selection, to come on stage.
"King, have you finished building the model of this Go project?" the chief judge asked.
"Yes! It's been done." Yuxing replied, "I wonder if this has anything to do with us getting the key to the computer room?" This is actually a disguised expression of dissatisfaction.!~!
{PiaoTian Literature www.PiaoTian.com thanks all book friends for their support, your support is our biggest motivation}