Guide
Huang Peihong with his smart baby Iron Egg No. 1
"Hello! Putian Evening News Hotline? "Yes. Hello! Is there anything you need help with? "That's right. My name is Huang Peihong, is your newspaper 20 years of loyal readers, I am an artificial intelligence amateur, I studied a smart baby called 'Iron Egg No. 1', please ask your newspaper is interested in interviewing and reporting? "Sure, we'll make an appointment for an interview." "Okay." ......
A few days ago, the evening news hotline 2516110 received a call from Putian reader Huang Peihong, reflecting his some achievements in artificial intelligence, which aroused the strong interest of the evening newspaper. Through the news hotline, after several contacts, the evening news immediately conducted an exclusive interview with Huang Peihong to discover the professional, persistent and achievement story behind his artificial intelligence research, so as to entertain readers and hope to attract the attention of relevant parties.
Not long ago, Huang Peihong, a member senior engineer of the Chinese Engineering Intelligence Society, conducted an academic discussion on Li Guojie's academic text "Some Cognitive Issues Related to Artificial Intelligence" in the 8th issue of the authoritative journal Computer Society Newsletter, and the relevant comments were recognized and accepted and published in relevant columns, which introduced Huang Peihong's research results and experience in natural language understanding. Li Guojie is an academician of the Chinese Academy of Engineering, one of the leaders of Chinese intelligent research, has been engaged in high-tech research of the National 863 Program for a long time, and has twice served as the chief scientist of the national "973 Program" project. Huang Peihong, who is unknown in mainstream journals, has published a number of insightful academic texts on artificial intelligence, his original research results natural semantic understanding theory began to be published in the mainstream journal China Computer Society newsletter, the papers he wrote about natural language understanding were included in the international authoritative computer science literature library dblp, and he was invited to Beijing, Shanghai and Putian University for many times to make academic reports on natural language understanding, and in 2020, Huang Peihong was also elected as a director of the Fujian Artificial Intelligence Society.
Putian Evening News: For most ordinary people, your theory seems esoteric and unfamiliar, what can you summarize this theory in one sentence?
Huang Peihong: The purpose of this theoretical research is to enable machines to think and understand, so that they can better solve problems and handle affairs.
Putian Evening News: Where is this theory unique than others?
Huang Peihong: This theory is based on natural semantics, through natural semantic understanding so that machines can really think about real understanding, which is different from existing technology, with originality. Today's systems or machines are relatively mechanical and have no understanding at all, and can only rigidly follow the steps and commands set by people step by step, and the cycle begins.
Putian Evening News: Who else in Putian, the province and the domestic side is doing this?
Huang Peihong: Putian experts in this regard include Chen Zhaoxiong, a well-known machine translation expert, who is now the chairman of China Electronics Technology Group and a former vice minister of the Ministry of Industry and Information Technology. He was a Ph.D. in the Institute of Computing, and he had written to me in the 90s and was encouraged. In Fujian Province, I know professor Zhou Changle of Xiamen University, and Shi Zhongzhi, a researcher at the Institute of Computing of the Chinese Academy of Sciences, who did cognition earlier in China, which belonged to the image thinking school advocated by the famous scientist Qian Xuesen, and my theory should also be classified in this school.
Putian Evening News: Why did you think of making this thing? How?
Huang Peihong: This matter can be traced back to the middle school era, I began to have a strong interest in artificial intelligence in high school, when I wrote a small paper and won an award in the city. In 1986, I had the honor of being admitted to the top computer department of Beihang In China, when the department had a strong atmosphere of artificial intelligence research, when I designed the graduation project, because I thought that recognition, especially image recognition, was the key to artificial intelligence, so I chose the image recognition topic, the tutor let me do the speech recognition topic, and put forward an idea that was praised by the defense experts. The failure of the fifth-generation computer program in the early 1990s caused me to think that they had not passed the image recognition test, had not yet reached the highest level, and began to carry out cognitive experiments following the three principles of introspection set by the psychologist Wundt himself. Usually, in the course of learning the course, I realized that understanding is very important, and knowledge must be understood in order to be integrated, and began to study the problem of understanding. Through checking the data, we know that many scholars have long realized the importance of natural language understanding problems, such as Ma Xiwen Yu Shiwen of Peking University in China, Huang Zengyang of the Institute of Acoustics of the Academy of Sciences, Dong Zhendong of the Academy of Military Sciences, etc., but there is still no clue about what understanding is. This problem has been thinking in my mind, once in the car to ponder this problem, remember the previous understanding of artificial intelligence, link all aspects, suddenly came inspiration, to find the key point of this problem, all at once figured it out, but after opening the "Ren Dou second pulse" also need to be verified, bold hypothesis careful verification and spent a lot of time. After several twists and turns, in 2007, I published a paper "Machine Cognitive Form System for Natural Language Understanding" in the main journal of computer engineering and science sponsored by the National University of Defense Science and Technology, which has been around for 15 years.
In 2020, I published the paper "Research and Development of Secure Intelligent Computers" in the computer professional journal Modern Computer, which is included in the Selection Database of Core Journals in China, and these two papers will lay the foundation for the research and development of general artificial intelligence. In 2010, he published the paper "Intelligent Definition Research Based on Semantic Origin Intervention" in the journal of mind and computing sponsored by the Fujian Provincial Key Laboratory of Brain Intelligence in Xiamen University, which demonstrated the equivalence of intelligence definition and Turing test through multimodal Turing test, which included visual Turing test, and then the internationally proposed visual Turing test paper was similar, but later than my article. The research results achieved over the years are inseparable from the exchanges and help of friends and students, and I have always been grateful.
My papers on natural language understanding were included in the international computer science bibliographic database "Exploration on Causal Law of Understanding and Fusion Linking of Natural Language" published in 2017 and the paper "Ten-Years Research Progress of Natural Language" published in 2018 Language Understanding Based on Perceptual Formalization", it is worth mentioning that the dblp database has a good reputation in the academic community, and its authoritative research community is highly recognized. Both papers were searched by EI. Two papers in this series were accepted for publication by international conferences and searched by the ei and istp databases, respectively. I was also invited to give an academic report and exchange discussion at the International Conference on Intelligent Science icis2017 and the International Symposium on Advanced Computational Intelligence and Intelligent Informatics iwaciii2017, and was invited by the School of Information Engineering of Putian University to give an academic lecture on "True Understanding of Natural Language and Next-Generation Intelligent Computer Models" in 2019. In 2020, I attended the annual meeting of the Fujian Artificial Intelligence Society and was elected as a director of the Artificial Intelligence Society, and was a member of the Chinese Engineering Intelligence Society.
Putian Evening News: As far as I know, you are usually quite busy at work, can you still make such a thing?
Huang Peihong: This thing is done outside of eight hours, whether there is time mainly depends on interest, interest has time, no interest will never have time. Within eight hours of business work, I performed my job duties well, I successfully developed the earliest intermediate business software of Putian Agricultural Bank, solved many difficult and complicated diseases in computer business, and some software development concepts were used by superior departments to learn from and apply, and I was one of the backbone of the department with strong technical ability and more workload. This research itself is also related to the computer major, and the research has a promoting effect on the business, and this research result can also play an important role in the design of the unit's digital transformation automation software in the future. In 2018, the relevant research results were published and won the Award for Glory awarded by the unit, and the relevant research was affirmed by the text evaluation of the scientific and technological personnel of the headquarters and was invited to give a video lecture and exchange.
Putian Evening News: To what extent is this thing currently done?
Huang Peihong: Theoretical research has been completed and entered the application development stage. According to the published records, in terms of theory, I have discovered the WHAT-why understanding effect in natural language processing, and through experimental verification, on this basis, I have established a machine cognitive form system of natural language understanding and run it through prototype program development and debugging, testing the correctness and validity of the theory. Breakthroughs have been made in the study of natural language comprehension issues. In terms of application development, the prototype system Iron Egg No. 1 has completed development and is ready to enter the development phase of Iron Egg No. 2.
Putian Evening News: What is the current general environment for research and development in this regard?
Huang Peihong: At present, the government still attaches great importance to the general environment and has issued relevant planning policies, such as the artificial intelligence 2.0 plan. The actual implementation of the Putian Municipal Government is still unclear.
Putian Evening News: Why is it called Iron Egg No. 1?
Huang Peihong: When the countryside gives names to newborn babies, they often take a cat and a dog, ask them the reason is that the name is "cheap" to feed, this system has a more earthy name called iron egg, iron egg series grows to 6.0 when basically has the mind of an 18-year-old adult.
Putian Evening News: What will this thing look like in the future? What is the vision?
Huang Peihong: The final appearance of this thing should be embedded device, that is, it is embedded where there is wisdom, for example, it is embedded in an electric vehicle, it is an autonomous electric vehicle, embedded in a machine tool It is a smart machine tool, embedded in the teaching software is intelligent education, embedded in the bank information system is an intelligent banking service system, etc., this is a general intelligent computer, equivalent to humans, and developed to the highest realm of brain-like computers, which are essentially the same. Today's brain-like computers have not yet reached the highest level, essentially only equivalent to the current computer, if not applied to the theory of natural semantic real understanding to expand it. I have now also arrived at the year of knowing the destiny, and strive to see the general intelligent computer can be successfully developed in my lifetime, the general computer is upgraded to a general intelligent computer, and the intelligent computer is widely used to become a reality, solving the productivity, so that the intelligent computer can become a good helper for mankind and benefit all mankind.
Putian Evening News: What is your best hope for the future?
Huang Peihong: The paper "Machine Cognitive Form System of Natural Language Understanding" involves the innate instinct mechanism of machine learning, which has given the truth of how common sense is obtained through learning, this multimodal common sense is a kind of natural common sense, which is equivalent to human common sense and has universality, so the problem of natural language understanding is solved, and the common sense problem is immediately solved. The model established, the model of the world, is equivalent to that of humans and is equally versatile. This paper is based on multimodal information for language understanding and processing, considering the rich world knowledge, and associating the world knowledge of different modes, that is, breakthroughs have been made in the problem of real understanding of natural language through multimodal correlation. Some experts have proposed that the problem of natural language understanding should be solved by solving the problem of common sense, and some experts say that there is still a lack of semantic understanding theory of multi-modal or multi-source knowledge association in China, and some people have only predicted that multi-mode may lead to a major breakthrough in natural language understanding. It is hoped that peers will jointly protect innovation by respecting originality, and push China's artificial intelligence research to a new level and a new height.
Finally, I would like to say a little suggestion to the government departments, please criticize and correct the deficiencies. I am a senior engineer majoring in computer science, but also a director of the Fujian Artificial Intelligence Society, research artificial intelligence for more than 30 years, the original research results on artificial intelligence have matured, according to the actual situation in Putian, in view of the great importance that Putian Attaches to the revitalization of science and technology, I take this opportunity to boldly suggest that Putian City set up a general artificial intelligence laboratory, and at the same time suggest that the relevant departments of Putian City can incorporate the real understanding of natural language research into the project support plan when implementing the State Council or Fujian Province's new generation of artificial intelligence planning. Ignite the flame of the general intelligence industry, gradually make the intelligent industry bigger and stronger, base itself on Putian, face the whole country, and make general intelligence a business card of Putian.
Source: Putian Evening News reporter Lin Shuanghui
Editor-in-charge: Chen Hengshan
Editor: Yan Shaoxiang