The challenge of brain science
Xinmin Weekly: Academician Yang, you are the founder and promoter of China Brain Project. Since 20 13, the United States, Japan and the European Union have also launched their own "brain plans". Why do you think countries around the world attach so much importance to the "brain project"?
Yang: This is because, on the one hand, the study of the brain is the most important challenge for us to understand nature and ourselves, because the brain is the headquarters that controls our whole body. How does the brain work? This is regarded as the so-called "last frontier" in the whole natural science field, that is, the "last frontier" After conquering this field, it seems that there is no other field to conquer further.
So what's the problem? We know about nature and the world through our brains. Now, it is a great challenge for us to know how our brain works and moves in it. For a long time, people even doubted whether people can understand the brain itself through our own brain, because there are profound philosophical problems in it. But now it seems that scientists have done a very good job in this regard. Their understanding of the brain and nervous system is progressing very fast, and there are many very interesting research results.
So, why is knowing the brain an extremely important challenge for us to know nature and ourselves? Because our brain is composed of about 86 billion nerve cells, which is roughly equivalent to the total number of planets in the whole galaxy, there is a connection point between nerve cells, and the number of this connection point is about trillions or even billions, thus forming a very complicated neural circuit, that is to say, a single cell is not isolated, but forms a connection circuit through such connection points, just like an electronic chip of ours, which is composed of various electronic components. Therefore, the brain is considered to be the most complex system, and this system is constantly changing. With the development of people, various external and internal conditions will change constantly, not only its function will change, but also its structure will change, so this is a very complex and ever-changing research object, and the complexity of such a research object is incomparable to other systems.
On the other hand, the brain and nervous system, like other systems in our body, will also have problems, such as Alzheimer's disease, Parkinson's disease, paralysis caused by spinal cord injury, stroke and other diseases, which will cause abnormalities in the nervous system, so this situation will become more and more serious with the aging of the whole society. For example, the population over 60 in Shanghai accounts for about 3 1% of the total population. Therefore, in this process, after the nervous system and brain are aging with human beings, the degree and quantity of senile diseases and nervous system diseases are incomparable before, and such diseases have brought heavy economic and spiritual burdens to society. In addition to senile diseases, there are also mental diseases such as depression, including autism in children, which are also related to the nervous system. These diseases have brought more and more heavy burdens to the whole society.
The prospect of artificial intelligence
Xinmin Weekly: Will artificial intelligence surpass human beings?
Yang: Our brain and nervous system are very effective organs for processing information. For example, we can see all kinds of colors and shapes, and we can react quickly. The working principle and operating mechanism of some brains will inspire us to develop artificial intelligence.
We have two ways to realize artificial intelligence. One way is that no matter how the brain works, we finally realize a function similar to our human brain in other ways. Another way is to learn from the working principle of the brain and see how the brain works to develop brain-like artificial intelligence, that is, to study artificial intelligence through some ways inspired by the working principle of the brain. Francis crick, a famous biologist who won the Nobel Prize, once said: "In a broad sense, for a scientist engaged in neuroscience research, a universal principle is that evolution is much smarter than ourselves."
Because our brain and nervous system have become the current working mode after long-term evolution, this way has become very effective and efficient because of evolution. This way is better than thinking hard by yourself and finally using other methods. This is a principle that we generally abide by, but it does not rule out that we have some new ideas that go beyond the evolution of our brains. Especially after I have a certain understanding of the working principle and operating mechanism of the brain and nervous system, I think it is really unique. It has its own characteristics and can form such a system with limited space, low energy consumption and effective information processing.
Judging from the whole evolution process of human beings, in fact, the truly intelligent people are about 654.38 million+years. In this sense, human beings are also in the process of continuous evolution, and the evolution of human beings has just begun, which has a long-term development prospect. So in my opinion, human wisdom is infinite and the future is insurmountable. I think artificial intelligence will be integrated into human society, but it can't surpass human beings in the overall sense and from the perspective of intelligence, because human intelligence is also constantly evolving.
As for a certain function, such as calculation, 1000 bits multiplied by 1000 bits, then no one can do it, and the computer has already lost. 1997, "Deep Blue" defeated the chess champion. I said, "We may still be competing with it now, but one day we will not win." The same is true of the man-machine battle of Go now, so I don't think it is excluded that some aspects of human intelligence are not as good as artificial intelligence. But in the long run, artificial intelligence will not surpass human beings, and it can only be gradually approached, but it cannot be surpassed.
Xinmin Weekly: According to foreign media reports, the artificial intelligence "Watson" developed by IBM in the United States was input by researchers into 20 million cancer research papers. Through systematic analysis, a 60-year-old female leukemia patient was diagnosed in only 10 minutes, and the condition was difficult to judge, and a suitable treatment scheme was put forward. 20 15 The United States also put forward the concept of "precision medicine". Will artificial intelligence play a big role in this regard when medical resources are in short supply?
Yang: I think the data and information accumulated in the past are of great significance to both human doctors and robot doctors. According to their own understanding of this problem and the accumulation of past experience, doctors analyze the indexes of patients, make logical reasoning and finally make judgments. This is the same. In this sense, artificial intelligence can really play a very important role, because the speed of receiving and processing information is relatively fast. In a sense, this is a bit like playing chess, because when playing chess, an experienced person has a lot of chess manuals in his mind. According to these chess manuals, he can predict what will happen after many steps.
But we should see that at least for now, it is impossible to create a new means, because medical technology and medical research are constantly developing, and both need human beings to input new things. Of course, for a human doctor, he also needs to receive new information, but the key is that people are creating things. A famous psychologist once said how to define intelligence. Intelligence itself seems empty. What is intelligence? He said that when you don't know what to do, you use intelligence.
This is a vivid and simple statement, but it is the truth. Now, it is difficult for robots to do this, so it will have its advantages and it will play a very important role in the whole medical practice. Artificial intelligence experts at home and abroad are also exploring the treatment system of artificial intelligence, but it has inherent limitations.
Xinmin Weekly: Apart from the use of intelligence you just mentioned, is there anything else that artificial intelligence can't imitate the human brain?
Yang: The relationship between doctors and patients is not like that between cold robots and people. He needs to communicate with people mentally and emotionally, which may be of great significance to some diseases. Such spiritual communication is not invisible, but actually mobilizes the patient's own immune ability, which cannot be completely replaced by drugs.
Why do some cancer patients, who were originally fine, suddenly collapse after learning that they have cancer? As a material basis, the body's defense system suddenly collapsed. In this case, he produced many cytokines that destroyed the immune system. He died soon after his immune system was damaged. However, if emotional encouragement is given to patients, through various exchanges, it will help to improve patients' own immune ability, improve the overall state of the body, and be more conducive to taking the initiative in the struggle against diseases, so I think this is a very important piece, and it is also impossible for artificial intelligence to imitate now.
Xinmin Weekly: In the future, can artificial intelligence even imitate human emotions?
Yang: What I want to say is: "Nothing is impossible." Scientifically speaking, it is much more difficult for you to say impossible than possible. I often give an example. When humans don't know X-rays, people will wonder how people can see the bones inside. Isn't this pseudoscience? This is unimaginable, but with X-rays, it becomes a natural thing.
For example, the sixth sense, so far, there is not enough conclusive evidence to show its existence. I even sometimes feel as if someone is watching me from behind. Looking back, I found that someone was really watching me. Of course, you can also say that this is a probabilistic event that happened by chance, so you can't say that people only have five senses. Do you feel anything else? Do you have telepathy? These are difficult questions to answer. Can't say unscientific. Can only say that the current science can not explain, more accurate.
Diligence is the most important thing.
Xinmin Weekly: In recent years, Super Brain, a large-scale scientific competitive reality show, became an instant hit. There is a player named Wang Feng, who has a strong memory and is admired by everyone. Some parents think that if they can have good memory ability, they will greatly improve their children's learning efficiency, especially in foreign language learning. Do you have any special training and memory methods in the process of language learning?
Yang: In fact, one of these people with super memory is the most famous person and the realistic prototype of the protagonist in the Oscar-winning film Rain Man. His name is Pickering. It is said that this man can recite about 7,900 books, and some people say that he can recite more than 9,000 books. One of his habits is to put the book on the shelf after reciting it. He is the only person in the world who can read two books, except. His brain magnetic resonance imaging showed that we normal people had a left hemisphere and a right hemisphere with a corpus callosum in the middle, but he didn't. He is flawless, but it is not clear whether this is the basis of his super memory. Even if this is the basis of his super memory, why is it seamless? Why is a whole the basis of super memory?
This is an example I often give when I give lectures on popular science. It is reported in the medical history from 65438 to 0789. This is a proper noun. Translation is "low intelligence and special talent syndrome". He knows a lot, but his IQ is very low, just like conductor Zhou Zhou. He is a patient with Down syndrome, but he has unparalleled talent in music command. King Pique is the prototype of Rain Man. In "Super Brain", there was a man named Leo Chou, whose number was marked by the root number several times. He is faster than an associate professor in the department of mathematics. It is said that once he was disobedient, his father hit him with a pillow, and then it became like this. This can't be proved, but Jiang Changjian, the host at that time, said he wanted to restore his reputation. People say that his IQ is very low. Look at him.
Let's go back to learning foreign languages. I'm not a "special talent syndrome with low intelligence". If I have any talent, it may be in learning languages. Of course, I like learning foreign languages very much, so interest is the best teacher. In this process, I constantly explore my own methods and rules. For example, the grammatical rules of foreign languages are actually similar. I studied Russian at first. After learning Russian to a certain extent, I think the whole grammar rules are basically the same. So after you have mastered all these things, try harder. I really learn a little faster than others, but most of us are not geniuses. This difference can be made up by your efforts the day after tomorrow.
There are also real geniuses, like Einstein, and I admire them. These people are naturally better than others. In this case, no matter how hard you try, you can't reach his level.
I haven't tested it myself. I define myself as above average, definitely above average, but I think the most important thing is my diligence. Diligence is not a day or two, nor a month or two, nor a year or two, nor even ten or twenty years. In fact, I didn't walk fast, but because I also walked forward when others stopped, in a sense, I walked ahead.
Therefore, diligence is a long and continuous process, not so easy. Actually, I'm not necessarily much better than others. Of course, there is also a question of opportunity. I have met many good opportunities, and some people have helped me, but the key is to do it well. Opportunity favors those who are prepared.