At the 2024 Inclusion Bund Conference, Professor Michael Jordan, the master of machine learning, once again shared his profound insights into artificial intelligence. He keenly pointed out that there are three key deficiencies in the current development of artificial intelligence: insufficient attention to collectivity, uncertainty and incentive mechanisms. The editor of Downcodes will interpret Professor Jordan's wonderful speech in detail and discuss how to better integrate artificial intelligence into the industry and achieve sustainable development.
“The lack of attention to collectivity, uncertainty, and incentive mechanisms are the three missing aspects in the current discussion of artificial intelligence.” On September 5, at the opening main forum of the 2024Inclusion Bund Conference, machine learning dean, U.S. Michael Jordan, "Academician of the Three Academies", once again brought the latest insights into artificial intelligence after a year. Michael Jordan believes that for artificial intelligence to be implemented in the industry, a collaborative group needs to be formed; to build an artificial intelligence collaborative system, the "incentive" perspective of economics must be introduced.
At the main forum of the Bund Conference, Michael Jordan once again talked about the uncertainty of artificial intelligence. "ChatGPT, are you sure what you just generated is correct?" He pointed out that current artificial intelligence systems have difficulty expressing what knowledge it has really learned, nor the ability to express how certain it is. In comparison, humans are faced with Excellent in times of uncertainty, especially when working together as a group to cope with it.
Therefore, Michael Jordan suggested that not only individual devices should have certain intelligence, but artificial intelligence should also be reflected at the overall system level through collaboration. He pointed out that it is not enough to integrate human intelligence into super-intelligent computers. The application of modern information technology in the fields of medical care, transportation, financial technology and business requires collective and decentralized intelligent systems.
Michael Jordan further explores the relationship between uncertainty and collectivity. He pointed out that humans can better cope with uncertainty when collaborating collectively, but how to make current AI systems have similar collective collaboration capabilities is still a key unsolved issue. He believes that the microeconomic perspective is a lack in current AI research.
"Incentive mechanism" is a key factor in market economy and collective intelligence. "AI has massive data, but some of it cannot generate value. Only by designing an incentive mechanism can AI agents be driven to contribute and collaborate." Michael Jordan proposed "three layers of data" Market (Three-Layer Data Markets)" model, in which users, platforms and data buyers form a closed loop through "selling data", "purchasing data" and "providing services". He emphasized that data buyers, that is, enterprises, can combine "data and services" to establish an incentive mechanism with users, thereby bringing them real value.
In this regard, Michael Jordan invoked statistical contract theory, a new theory that combines statistics and economics. In contract theory, the agent owns private information, and the principal forms a market in which data and services promote each other through incentive mechanisms, maintaining a balance of interests between supply and demand.
For example, airlines are divided into "business class" and "economy class". As a principal, the airline can provide different prices according to the agent's different willingness to pay without requiring the agent to disclose his personal information. As regulation of data privacy has increased globally over the past decade, he also suggested that “we can further improve user utility through non-uniform privacy requirements, imposing higher requirements on low-cost platforms.”
Artificial intelligence, an emerging field of engineering, is connecting humans in innovative ways through large-scale systems. Its development is similar to the rise of chemical engineering in the middle of the last century and electrical engineering at the end of the 19th century. The former is based on chemistry, fluid mechanics and other fields, and the latter is based on electromagnetics, optics and other technologies. Artificial intelligence systems are based on the past 300 years of human reasoning concepts, algorithmic concepts, and economic concepts, and need to target human welfare. Michael Jordan cautions, “But AI’s rise and development are being distorted by being framed by ill-thought-out, homespun old-fashioned visions.”
Professor Michael Jordan is a pioneer in the field of machine learning. He established the mathematical and computational foundation for machine learning by establishing connections between machine learning, probability, statistics and graphical models. He has received the IEEE John von Neumann Medal, the International Joint Conference on Artificial Intelligence Research Excellence Award, and the 2022 First World Association of Top Scientists Award.
The 2024Inclusion Bund Conference was held at the Shanghai Huangpu World Expo Park from September 5th to 7th, with 1 opening main forum and 36 open opinion forums. Recently, the authoritative media "Asia Science and Technology Daily", which has long been paying attention to global technology trends, selected the four "most anticipated global innovation and technology conferences in the second half of 2024", and the Bund Conference was selected.
Through Professor Michael Jordan's wonderful sharing, we have a clearer understanding of the future development direction of artificial intelligence, and we also see the challenges and opportunities it faces. It is expected that artificial intelligence technology can better serve mankind and benefit society in the future.