During the conference, the " Statistics and Data Science Forum in the Era of Artificial Intelligence
"
was held
. The forum was chaired by Chen Songqi, chair professor of Peking University and academician of the Chinese Academy of Sciences. Su Meng, chairman and CEO
of Baifendian Technology,
was invited to participate.
, Peter Bühlmann, current chairman of the Institute of Mathematical Statistics (IMS), Jiang Song, deputy director of the National Natural Science Foundation of China, academician of the Chinese Academy of Sciences, Ma Zhiming, researcher of the Institute of Applied Mathematics, Chinese Academy of Sciences, academician of the Chinese Academy of Sciences, professor of Xi’an Jiaotong University, academician of the Chinese Academy of Sciences Xu Zongben, Princeton University of Frederick L. Moore'18 Title Professor of Finance Fan Jianqing and other experts discussed the cultivation of statistics and data science talents, as well as how to deal with the opportunities and challenges of the artificial intelligence era.
At the meeting, experts and scholars pointed out that statistics and data science are the cornerstone of data resource development and have become an important part of modern society. With the development of the times, statistics and data science are facing new opportunities and challenges.
Ma Zhiming said that we are currently in the era of big data, which has promoted the development of data science, and data science has developed into an independent discipline that intersects with multiple disciplines. At the same time, the massive nature, heterogeneous nature, and non-IID nature of big data also bring great challenges to traditional statistics.
In this regard,
he believes that we urgently need to develop and innovate statistical ideas and statistical theories, establish a statistical foundation for big data analysis, and establish a rigorous statistical analysis system and calculation foundation for data science.
Xu Zongben pointed out that ChatGPT has triggered a wave of AI large-scale models, and artificial intelligence research has entered the era of large-scale models (AIGC) from the era of deep learning, and has shifted from focusing on special artificial intelligence to general artificial intelligence. The essence of the big model is the learning methodology, and the essential requirement of the learning methodology is to seek the integrated design and control of data, network, loss, algorithm, task and other elements. In this regard, he proposed the mathematical framework, strict definition, mathematical model and general algorithm of SLeM, and shared its successful application in multiple machine learning automation problems.
Fan Jianqing believes that whether the development of statistics can be recognized by the society depends on whether statisticians can solve its needs for the society, whether they can serve as scientific ministers, learn the language of other disciplines, and refine the problems of other disciplines into statistics. Problems, while solving statistical problems, help other disciplines solve their problems. He said that regardless of whether statisticians participate or not, statistical science will continue to advance with the needs of society and the development of artificial intelligence. The key is whether statisticians can grasp the current opportunities.
Su Meng said that statistics and data science are the basic supporting disciplines of the digital economy. In the process of industrial digitalization, enterprises need three types of data science talents: data engineers, data analysts and data scientists, as well as software engineers, system architects, domain experts and Business analysts, etc., to meet different business needs such as business requirement understanding and translation, model architecture design and tuning, data governance and processing, etc.
In terms of data science talent training, Su Meng emphasized three points: First, horizontal capacity expansion requires large-scale multi-source heterogeneous data processing capabilities, as well as large and small data fusion processing capabilities; second, the combination of statistics and business fields, With the help of statistical advantages, the data can be better used; the third is the deep integration of industry, academia and research, and the use of new generation information technologies such as big data and artificial intelligence to continuously improve productivity, realize the implementation and transformation of technological innovation, and promote the development of the entire society.
Percent Technology will continue to be oriented by
solving major practical problems, focusing on the innovative practice of data science theory and technology, realizing the implementation and transformation of scientific and technological innovation, promoting the deep integration of industry, education and research, and helping China's statistics and data science research and application to reach new heights steps.
