Chapter 228: [Nov / Navy]

, The latest chapter in the latest global monopoly of technology!
Luo Sheng returned to the company's headquarters and started a new plan in his office where Bluestar Technology is located. I am afraid that only him can complete this plan, and he only wants to do it by himself.
The development of AI requires supercomputers. Luo Sheng does not currently have a personal supercomputer. The two companies also do not have their own independent supercomputers, unlike IBM.
However, Luo Sheng now has a data center and cloud computing, and the underlying layer of cloud computing is also a supercomputer. Through the data center and operating system, ultra-large-scale physical servers in different regions are aggregated into a supercomputer for unified scheduling and use.
No one on either side is saying that because they have their own strengths and their priorities are different, and with the development of technology, cloud computing and supercomputing will also become you, me and you, together for big data. Times provides powerful technology engines.
...
At this moment, Luo Sheng created a new folder on one of his working computers. The index directory is called "Nov". This is the code name or name of the artificial intelligence to be developed.
According to the pronunciation of Nov, and then transliterate it with its homophonic, the Chinese name is "Na Fu".
After determining the name, Luo Sheng began to program and develop Nafu's agent.
If you want to develop artificial intelligence, you must first understand three concepts: artificial intelligence (AI), machine learning (ML), and deep learning (DL).
These three are closely related. In simple terms, artificial intelligence is to give a device or a machine the same intelligence as humans. Machine learning is a method to implement artificial intelligence. Deep learning is a technology to implement machine learning.
Artificial intelligence includes machine learning, and machine learning includes deep learning.
Last year, that is, in 2006, Jeffrey Hilton, a professor of machine learning in the University of Toronto, and his student Ras Lahudinov published an article in Science that opened the depth. The wave of learning in academia and industry.
...
In the following days, Luo Sheng established an end-to-end "ecosystem" for the development of Nafu, and wrote programs for him.
Recently, Luo Sheng did not care much about the two companies, and started to work as a shopkeeper for a while. On the one hand, the company is not busy now, and it has been arranged clearly. Secondly, there is Yao Jianhong on the Blue Coast and Zhang Bowen on Blue Star Technology.
This "ecosystem" is currently temporarily arranged in the underground vault of his private villa, waiting for his super mansion to be completed. After the overall relocation, Luo Sheng has now spent billions on creating a sense of technology. A super-luxury mansion is located on the shore of Yangcheng Lake.
The development of artificial intelligence does not happen overnight. This is a continuous upgrade and optimization path, from simple programs to weak artificial intelligence, strong artificial intelligence, and even super artificial intelligence.
The upgrade includes advanced software and hardware. Instead of typing a few lines of code on the keyboard, a smart AI with self-emotional logic analysis is born.
The "ecosystem" created by Luo Sheng includes designing a simulator, a programming language, an integrated programming environment, an algorithm library, a set of tools for applying deep learning to firmware, a set of teaching logic courses, and Cloud computing support.
Just so, this "ecosystem" is not enough.
To achieve true strong artificial intelligence, more advanced and more hardware support is needed, so as to create the ability to simulate the perception, action and cognition of the human brain. It can use complex structures or multiple nonlinear transformations. Multiple high-level processing of data.
Obviously, deep learning algorithms are extremely critical.
The other key link is data. Only companies with large amounts of big data have the advantage or foundation of developing artificial intelligence.
There are actually not many companies or institutions with such advantages in the world. There are Google, Yahoo, IBM, etc. abroad, and there is no doubt that they are "L-BAT" in China, especially "L", that is, Blue Star Technology Group.
The amount of data it holds can be said to be among the best in the world. BlueSpace BlueSpace, Yunge Search, Reddit, Bluestar Video and so on. The business segments of Bluestar Technology are full of vast data.
However, the "Nov-Nauf" to be developed by Luo Sheng is not an asset of Blue Star Technology or the French Riviera, and is completely independent. It belongs to Luo Sheng's personal proprietary technology, but will provide AI technical support for these two companies in the future. It is certain.
As the founder of Bluestar Technology Group, Luo Sheng Group's database is undoubtedly completely open to him. The forthcoming "Nov-Nav" will enter the database of Bluestar Technology Group and the French Riviera.
...
Time passed day by day, and Na Fu's prototype also completed the pioneering stage under Luo Sheng's continuous stay up all night. Although it is far from the artificial intelligence in the true sense, this is a breakthrough of 0 and 1.
Regardless of this world, we have to accept that there is really a genius. People are more mad than people.
Even more annoying is that geniuses are working harder.
Luo Sheng came to the company early this morning, but he didn't deal with the company's affairs. He stayed in the office and was tossing Nafu. The original code had been written.
Luo Sheng opened his dedicated work computer to link to the cloud, and his private cloud service is located in the Sioux City data center.
The index quickly reached the "Nov" folder. After confirming the operation, it was started without any hesitation and awakened "Nav".
Watching the percentage progress bar continue to advance, Luo Sheng waited patiently.
However, he also knows that even if Nave is awakened now, it is also an
artificial mental retardation
. It must be taken step by step, from
artificial retardation
to
artificial intelligence
and then to
strong artificial intelligence
or The final ultimate form of "super artificial intelligence" was not achieved overnight.
Completing the first step is often the hardest.
[100%, startup is complete! 】
"Nice!"
Staring at the screen, Luo Sheng saw a sudden surge of energy in the scene. He said that he had already done a "plug-in" long before he said that he would immediately connect to the Internet.
At this moment, Luo Sheng opened Blizzard Battle.net, and quickly registered a Battle.net ID. The ID nickname was "Nov".
He then spent a few minutes arranging what Nov was going to do today, that is, letting her "shock" in the StarCraft: Brood War battle ladder system at Blizzard.com.
Compete with those top players.
This is Luo Sheng's first learning course for Nafu, allowing "she" to promote deep learning research in real-time strategy games (RTS).
StarCraft is essentially a fast-paced war game. Players play the role of commander in overall control, so any operation, decision, and consciousness of the player will directly or indirectly determine the outcome of a confrontation.
Players need to always consider the balance between economic and military development, and make a large number of decisions in a very short period of time, such as violent soldiers, mining, technology climbing, offense, defense, sneak attack and anti-sneak attack, harassment and counter-harassment, reconnaissance and counter-scouver Wait.
Such a real-time strategy game is undoubtedly an excellent platform for Nov to advance, but Luo Sheng gave Nafu some restrictions before she entered the celestial system to compete with those top human players.
The first is the fog of war, which is exactly the same as a human player, otherwise the map is hung up.
The second is to restrict the apm and epm of Nov. If there is no restriction, human players will lose 100% and it is impossible to beat. This is obviously not a fair confrontation, nor is Luo Sheng's original intention, nor is it a deep learning training for Nov. Great help.
The so-called apm is the number of operations per minute by the player. Like the top star players in South Korea, the apm can soar to more than 500 on average, and the epm is the effective number of operations per minute by the player. The apm is counted as long as the player clicks the mouse and keyboard Among them, blind points are counted, so apm has a large number of waste operations, and each operation of epm is an effective operation, such as repeated operations, this behavior is counted only once.
Luo Sheng set the apm upper limit of 550 for Nafu, which can achieve the same value as the world's top star professional players, while the epm on-line is set at 188.
At the operational level, Nafu is restricted to be among the top human players in the world. Then "She" needs to play at the "intelligent" level if she wants to defeat the top human players. To put it bluntly, consciousness is very high. important.
After Luo Sheng got everything done, he shook his spirit and decisively put Nafu into the battle network, allowing "her" to enter the celestial system to match the player to confront.
Blizzard's Battle.net system in Luo Sheng's eyes is that exists ~ EbookFREE.me ~ has 100% absolute confidence to ensure that the "glass slag" system cannot detect any anomaly in Nuff, so she pretends Adult players compete against real human players.
At the same time, Nafu has also access to Bluestar's database. Through deep learning algorithms, she continuously contacts and analyzes massive amounts of information. This does not prevent "her" from competing with human interstellar players on the ladder at the same time. .
This is the incomparable advantage of human beings compared with AI.
Next, Luo Sheng didn't do any intervention, sitting still, staring at the computer screen.
Now his computer has been taken over by Nafu, and enters the ladder system to start matching players. The competition mode is the 1V1 confrontation mode with the highest level of competition.
Luo Sheng didn't say a word at the moment. He quietly watched Nafu's first point of view. It took him less than ten seconds to search for his opponent and quickly found the player.
The first match officially began.
...