Artificial intelligence Nowadays, you are recognized as omnipotent, and the country is also keenly concerned about the development of artificial intelligence. The government also wants to support a local company and plans to catch up with the Korean home in a few years. Last year, there was a question about whether artificial intelligence could resurrect the "planned economy", and China's major giants had a heated debate.
In July this year, the State Council issued the "New Generation Artificial Intelligence Development Plan", and proposed the guiding ideology, strategic goals, key tasks and safeguard measures for the development of China's new generation of artificial intelligence in 2030, and brought China's development of a new generation of artificial intelligence to the highest level. Strategic point. In recent years, the rapid advancement of big data, cloud computing, and artificial intelligence (AI) has continuously enhanced human confidence in forecasting, planning, planning social and economic behavior.
Last year, on the issue of whether artificial intelligence could resurrect the “planned economyâ€, Ma Yun (microblogging), the main founder of Alibaba Group, and Qian Yingyi, dean of the School of Economics and Management of Tsinghua University, had a debate. Ma Yun’s point of view is that the market economy has been very good for more than a hundred years. His personal opinion is that there will be great changes in the next 30 years, and the planned economy will become bigger and bigger. The invisible hand of the market may be discovered. With the advent of the era of big data, the planned economy and the market economy have to be redefined. In the era of the Internet of Everything, the ability of human beings to acquire data far exceeds human imagination, and human understanding of the world will rise to a new height. Data will make the market smarter and make planning and forecasting possible.
In the economics and AI worlds, this debate has not stopped. In the future, the planned economy will “become bigger and bigger†as Ma Yun said, or will it be defeated as Qian Yingyi thinks? Some time ago, at the 2nd Yesanpo China Economic Forum, Xu Chenggang, a well-known economist and professor of economics at Cheung Kong Graduate School of Business, gave his own answer on the technical level and the market economy level.
Today, we are trying to develop artificial intelligence technology, how to develop? Is it coming from the plan? Still coming from somewhere else? This is the first question. Where will we bring us when artificial intelligence develops? Will it be brought to the planned economy? This is the second question. In order to discuss this topic, I want to start from the most basic technical level, understand what artificial intelligence means, what artificial intelligence can do, what can't be done, and what big data has to do with it.
When big data becomes a basic resourceFirst of all, big data itself is not that important. Its importance is that it is the foundation of artificial intelligence. Nowadays, China, Russia and all the developed countries in the world pay close attention to the development of artificial intelligence. The reason is that it is now clear that this is an emerging industrial revolution. The consequence of this industrial revolution is that a large number of unmanned factories will be produced, and many industries will become unmanned service industries. The high efficiency that has never been seen in human history will occur, and it will lead to a large number of human unemployment.
Since the development of artificial intelligence is based on big data, big data is now a basic resource like raw materials and energy in human history, but the difference between this resource is that it is not originally in the world. , but we manually collect it.
When this kind of production mode changes basically, will this new automation and comprehensive automation change the system from the fundamentals and bring hidden dangers? We need to learn from the industrial revolution that has happened in the past, otherwise it will repeat the same mistakes. The reason why the industrial revolution of the past brought lessons is because when these industrial revolutions were born, people overestimated where the industrial revolution might go, and when they overestimated their own power, they also abused these emerging sciences. technology.
Let me give you a few historical examples. During the second industrial revolution, a system like the central plan based on state-owned system was popular because it overestimated people’s planning ability and overestimated the ability of people to rule. Why can you make it? Another example is the destruction of the environment, such as the huge water conservancy project, thinking that "people must win the sky." How can people win the day? Because people's abilities are overestimated, people think that people can make the world's largest artificial lake. People can control the flow of water, thinking that people can change the environment according to people's will. But in fact, twenty years ago, there was a general consensus in the international community that the damage caused by the artificial lake to the environment is often not expected by us. Its damage often exceeds the benefits it brings, so it is internationally It is generally discouraged to recreate huge artificial lakes and large dams. Then there are fossil raw materials, and the large-scale use of fossil raw materials, along with the first and second industrial revolutions, has caused global carbon emissions and a series of pollution today.
These are the lessons of the past. Today, when big data and artificial intelligence are combined, there are many dangers that we may not know. For example, the government or a large company with a monopoly nature may use the data in its hands to try to control the society. Large-scale war, crime in this way, and so on.
The basis of artificial intelligence is measurable dataBy understanding the technical foundation of big data itself, we can understand what artificial intelligence can do and what it can't do.
First, the foundation of big data or raw big data is sensors and mobile devices. They first detect some specific data and send it through the Internet and the Internet of Things to be centralized. Therefore, the core of big data lies in data collection, transmission, storage and processing. All the data that sensors and mobile devices can measure is the key, what AI can do and what not to do is determined by these measurable data. The so-called deep learning artificial intelligence, its technical basis is to use big data to train the machine, to generate the ability to identify, the ability to reason, the ability to plan, so artificial intelligence includes planning.
Then there is the algorithm, deep learning is actually an algorithm, which is produced together with the decision theory in economics. What is the core of the algorithm? As an artificial intelligence device, such as a robot, the designer has to assign a purpose to the robot, which is what the machine is doing on the market? Its purpose is to discuss with economists one thing. It seeks to maximize its own benefits, or to maximize its own interests, and that is its purpose. But no economist really knows what the real purpose of the world is and what affects you. In abstract terms, people's goals are for happiness, for happiness, etc., but what affects your happiness, and what affects your happiness, no economist knows. This is why the market is wonderful and it is achieved by people on the market.
On the contrary, the purpose of artificial intelligence, if there is big data and artificial intelligence, is it possible to calculate or simulate its purpose through big data? This requires us to analyze human intelligence and artificial intelligence.
Human intelligence arises from the physical and psychological perceptions of human beings, as well as the information that people collect. As early as the 1950s when the economist Simon Simon discussed artificial intelligence, he had already proposed the concept of "recognition", which is the core concept of artificial intelligence today. The identification is divided into cold recognition and thermal recognition. Cold recognition can be understood as the recognition of the machine. Thermal recognition is the identification of people with emotions. The identification machine with emotions can not learn. There is also a very important basic concept, hard data and soft data. Hard data is all the data that can be measured and passed. What is soft data? There is no way to measure with the sensor, if it can not be measured, it cannot be passed. So when we talked about artificial intelligence based on big data training, it has technically determined that it has no basis for thermal identification and soft data. Does it even go to school on the basis of what? This is why the machine is not a human.
Biological science tells us that sense of smell, taste, and sexual desire cannot be measured. People's psychological perceptions such as joy, boredom, pain, thoughts, nostalgia, greed, and ambition are also unmeasurable. Another basic part of human intelligence is intuition. Intuition is a highly abstract, leaping response based on the combination of hard and soft data, cold recognition and thermal recognition. This kind of intuitively dependent data is not only unmeasurable, untransferable, but the intuition itself is beyond human description. There is a gap between human intelligence and machines. The basis of artificial intelligence is the data that can be measured, described, and transmitted. Training artificial intelligence to meet these conditions.
Therefore, deep learning such artificial intelligence does not ultimately involve learning the original basic elements of human intelligence. The reason is that it lacks the basic perception of a large number of people. When it does not have this basic perception, this machine is not likely to be similar to human beings through learning. The purpose can only be assigned by the person who set it.
Artificial intelligence cannot achieve national welfareThe economist always has a dispute about what the real effect of man is, and we never know. Therefore, the effect that a person assigns to a machine, it is not likely to be a universal function of human beings, it can only be defined and static in a narrow range. So in a broad sense, the objective function produced by any artificial intelligence device or robot is not and cannot replace the real person's own goal. When Simon won the Nobel Prize in Economics, he proposed a very important basic concept of "limited rationality", which has always affected the development of economics today, and is the most forward-looking thing in the development of economics today. Limited rationality is the problem that we realized when we discussed the plan. We are always one-sided. You assign an effect function to the robot, and it will not do better than you.
Therefore, the concluding observation is that you have no way to train the target behavior that arises from the original preferences and animal nature of the person. The artificial intelligence we can see today that can be learned in depth is actually limited to training the behavior of imitating people in known environments. Because you are trained by data collected in a known environment, such as consumer behavior in the market, the social behavior of those who participate in the discussion in a free environment, and, for example, facing a whisperer or a group of acquaintances. , the performance of the musicians. The behavior you collect is actually limited. It is limited to the existing system. You use this to train the robot. The robot will imitate what it is like in this state. Once it is out of the training environment, there is no original power. There is actually no way for artificial intelligence.
What is said above is what artificial intelligence can't do. Here is what artificial intelligence can do. Artificial intelligence can be planned or executed, but it can be planned and executed on the premise that the goal must be clear. Where will the revolution of artificial intelligence come from? It produces in all areas of the mission that you can imagine to accurately define it. For example, playing chess is complicated to calculate but the goal is simple. However, it is not so simple for humans to encounter a lot of things. It can be a very good assistant, assistant researcher, financial analysis assistant, doctor assistant, paralegal, military staff assistant, and so on. Why are all assistants? Because it has no way to replace people.
In general, if there are clearly defined and narrow targets, the robot will do it, including war. Of course, this is a serious controversial issue. In the most optimistic situation, artificial intelligence can ultimately plan and execute corporate and military tasks. Here back to the theme of economic analysis, military missions and economic missions are essentially different, why? Because the purpose of the military mission is simple, it is to win the battle. The purpose of the economic mission is unclear. The task of the economy is national welfare, not GDP growth. What is national welfare? National welfare is determined by the feelings of all the people. Even our people are confused, and the machines are even more unclear, so it is impossible to use machines to achieve them. Corporate governance and the national economy are fundamentally different, because companies pursue profits, and the national economy pursues national welfare.
Eliminating the market is tantamount to eliminating the foundation of the plan.Is it possible that the planned economy will replace the market economy after the development of artificial intelligence in the future? The answer is very simple. I have just made the basis of artificial intelligence clear. Big data comes from the market. If the market is eliminated, the data will be gone. You said, I have collected countless data that I have never seen in human history. I can stop the market and plan. Then I am wrong. Because after you eliminate the market, your foundation will be gone.
And the most important point, when the market is eliminated, if you try to use artificial intelligence, big data to solve resource allocation, it must be wrong. why? Especially in the allocation of resources related to innovation, a lot of work of these resource allocations is done by venture capital experts in the market, because they have a large amount of soft data, which can be judged by their intuition. Artificial intelligence can only process hard data, and there is no intuition so this judgment cannot be made.
Where are the talents of artificial intelligence today? As of the first quarter of this year, there are 1.9 million artificial intelligence talents worldwide, including 50,000 in China, and less than 40% of these 50,000 people have done this in 10 years. There are 850,000 people in the United States, and in the 850,000 people in the United States, 71.5% of them have more than 10 years of experience. The reason is very simple. In the market environment, there are so many people with innovative ability to come out. If the market environment is eliminated, innovation is difficult to develop.
There is a very deep relationship between big data, artificial intelligence and institutions, and the system will profoundly affect the development of big data and artificial intelligence. Because the collection and processing of big data will be restricted by the system, for example, which data collection is legal, whether it is supported or socially opposed, such as whether technology or monopoly companies have violated privacy rights.
On the other hand, the development of artificial intelligence itself is also restricted by the system. For example, in developed economies, when artificial intelligence is to be developed, a series of industries must be eliminated. How does the system itself face this problem? Another point is that a social equality or inequality can have a huge impact on the development of artificial intelligence. The reason is that the development of artificial intelligence will cause huge inequalities. A more equal society will be better able to solve this problem. In an unequal society, it will cause very sharp social contradictions and hinder the development of artificial intelligence.
The fan motor – an important component in industrial facilities.
Fans and their motors have become indispensable in modern industrial plants. Whether for ventilation of warehouses, cold stores for cooling or just for fresh air – fan motors are everywhere. The various fans available hardly differ in terms of function, but rather in their area and in the area of use. It is easy to recognize that a fan for cooling must meet other requirements as a fan for air-conditioning systems. Further attention should be paid to the control and energy efficiency. For new regulations, for example by frequency inverters, the energy consumption is reduced, and the life of the fan motors is also prolonged.
Fan Motor,Bathroom Exhaust Fan Motor,Cooler Fan Motor,Hvac Blower Motor,Blower Fan Motor,Electric Fan Motor
Shenzhen Maintex Intelligent Control Co., Ltd. , https://www.maintexmotor.com