Near the third generation of artificial intelligence ! Knowledge, data, algorithms, computing power are indispensable.
When artificial intelligence has gone through the outbreak of technology and entered a new era of landing applications and creating value, the importance of AI empowering traditional industries has become increasingly prominent. In the past, AI scientists were often asked about the changes that artificial intelligence can bring to people's lives. Now, they are thinking more about how deep learning-based artificial intelligence technology can be loaded into human consciousness to achieve trusted AI.
On July 9, academicians and top scientists from the field of artificial intelligence at home and abroad talked about a new paradigm of next-generation artificial intelligence at the 2020 World Artificial Intelligence Conference, focusing on and exploring the most cutting-edge technology development trends of artificial intelligence, focusing on artificial intelligence algorithm innovation , Multidisciplinary cross-fusion, and other topics, depicting the technological development path in the fields of intelligent vision, intelligent medical treatment, and intelligent chips.
AI has to do the same thing.
2020 is destined to be an extraordinary year. Hong Xiaowen, Dean of Microsoft Research Asia, said:
"In the process of jointly responding to the epidemic, gradually restoring production, and rebuilding social life around the world, digital technology represented by artificial intelligence is playing an increasingly important role."
The sudden emergence of a new coronary pneumonia epidemic has made the field of medical and public health a global focus. How to deal with the disease and the social problems caused by it and the global economic austerity has challenged everyone. It turns out that artificial intelligence technology has responded quickly. Joshua Bencio, a professor of computer science at the University of Montreal, one of the three giants of machine learning, said:
"Digital tracking technology can already predict the infectious risk of the new crown epidemic by gathering a large number of tiny clues."
In China, such as intelligent service robots, intelligent CT equipment, etc. are used in sheltered hospitals and other scenarios to achieve zero-touch diagnosis, intelligent escort and other functions to effectively control the infection rate of medical staff. As Microsoft Senior Chief Research Manager Katsushi Ikeuchi said:
"The biggest change that artificial intelligence has brought to people's lives is that this virtual-real world combines perspective and critical perception."
However, Zhang Cy, an academician of the Chinese Academy of Sciences and dean of the Institute of Artificial Intelligence of Tsinghua University, said:
"The current artificial intelligence only solves the problem of certainty in a complete information and structured environment. Although it has greatly promoted the application of AI, there are unsafe, unreliable, and unexplainable issues in environmental awareness, understanding the environment, AI security, and inference decision-making."
In addition, the academician of the Chinese Academy of Sciences He Jifeng also said:
"Because the data itself is biased, the unfairness of the mathematical model based on machine learning algorithms in decision-making is also a topic of concern."
Standing at the new starting point of artificial intelligence, Zhang Cy suggested that we must make full use of the four elements of knowledge, data, computing power, and algorithms to allow machines to learn in depth, and to adapt to random changes and inferences, in order to continue to advance artificial intelligence .
Multidisciplinary integration is the key.
Artificial intelligence empowers all kinds of things. When the world turns its attention to the third generation of artificial intelligence, a group of people are working silently to launch an impact on the basic theory of artificial intelligence and continue to study to break through the existing limitations.
Turing Award winner and academician of the Chinese Academy of Sciences Yao Qizhi believes that neural topology, privacy protection learning, and controllable SuperAI are the new directions of artificial intelligence theory research. Tao Dacheng, a professor at the University of Sydney and an academician of the Australian Academy of Sciences, is conducting research related to analyzing neural networks starting from the capacity and complexity of artificial intelligence models. Martial Hebert, director of the Robotics Institute at Carnegie Mellon University, described how to use intelligent vision to quickly estimate and speed up the judgment process.
From theory to algorithm, what else can we do for AI development? Dai Wenyuan, the founder and CEO of the fourth paradigm, believes that the current development of artificial intelligence faces three challenges: the first is the high threshold and the small number of data scientists; the second is the lack of data privacy protection technology, and the available high-quality data is still scarce; The third is the sharp increase in the cost of computing power brought by the development of AI.
Ramin Zabih, a professor of computer science at Cornell University in the United States, also said: "Artificial intelligence researchers are an inseparable, globally connected community, and the common interest of using technology to change lives will unite everyone."
Luc Van Gool, a professor at the Swiss Federal Institute of Technology in Zurich, believes that for deep networks, we are using many standard architectures, but a more reasonable way is that we should design an architecture optimized for a specific application.
In this regard, the president of Sapporo City University Hideyoshi Nakashima said:
"The development of artificial intelligence technology requires multidisciplinary integration."
Tang Xiaoou, a professor at the Chinese University of Hong Kong and founder of Shangtang Technology, said:
"Artificial intelligence is a very inclusive academic field, not only across the academic research field, across the industrial application industry, but also across the academic and industrial boundaries. The most important thing is that artificial intelligence crosses borders and requires international cooperation."
Thinking about the future, Kyoung Mu Lee, director and professor of the Institute of Artificial Intelligence at Seoul National University in South Korea, expressed his expectations for AI in the coming years:
''Hopefully more practical results will be published in the areas of medical treatment, autonomous driving, robotics and personal assistance. I hope to use AI to solve the problems facing humanity like environment, energy, aging and mitigation. In conclusion, AI can bring a high quality of life for us."
Social Plugin