IMCAFS

 Home

wang feiyue: research progress and prospect of generative counter network gan

Posted by millikan at 2020-02-27
all

Professor Wang Feiyue received his Ph.D. in computer and system engineering from renssellier Institute of Technology (RPI) in 1990. Since 1990, he has successively served as assistant professor, associate professor and Professor, director of robotics and automation laboratory and director of Advanced Research Center for complex systems at the University of Arizona. In 1998, he returned to China as a national Planning Commission's "plan for introducing overseas outstanding talents" and the "hundred talents plan" of the Chinese Academy of Sciences, and in 2011, he was the first national special expert of the "thousand talents plan" in the field of national defense. He used to be deputy director of Institute of automation, Chinese Academy of Sciences. Now he is director of State Key Laboratory of complex system management and control, Institute of automation, Chinese Academy of Sciences, director of military computing experiment and parallel system technology research center, National University of Defense Science and technology, director of China Economic and social security research center, Chinese Academy of Sciences, and President of Qingdao Intelligent Industry Technology Research Institute.

Abstract: the generative adversarial networks (GAN) has become a hot research direction in the field of artificial intelligence. The basic idea of Gan originates from the two person zero sum game of game theory, which is composed of a generator and a discriminator. It is trained by the way of antagonism learning, in order to estimate the potential distribution of data samples and generate new data samples. In the fields of image and visual computing, speech and language processing, information security and so on, GaN has been widely studied and has great application prospects. The main contents of this report include the background of Gan, the theory and implementation model of Gan, the development trend and the work we have done, that is, Gan and parallel intelligence. We think that Gan can deepen the idea of virtual real interaction and interactive integration of parallel intelligence, especially the idea of computing experiment, which provides a very concrete and rich algorithm support for ACP (artificial system, computing experiment, parallel execution) theory.

Previous highlights (click to view):

John Edward Hopcroft, academician of the American Academy of Sciences and Turing Award winner: deep learning

Professor of Nanjing University: Zhou Zhihua -- deep forest: exploring methods beyond deep neural network

Li Feifei, director of Artificial Intelligence Laboratory of Stanford University

Professor of Naval Engineering University: Ben Kerong: artificial intelligence technology and its application progress

Tsinghua University: Liu Yang -- machine translation based on deep learning

Professor of National University of Defense Technology: Yin Jianping: the past, present and future of computer science theory

Liu Yunhao, Dean of the school of software, Tsinghua University

Masashi Sugiyama, director of advanced intelligence research center of Japan Institute of science and chemistry

"Character feature" Deng Zhidong, Tsinghua University: "feature extraction + reasoning" small data learning is the key to the rise of AI

Minglue lecture hall Professor Ma Shaoping of Tsinghua University explained in detail "what can artificial intelligence do?"

[original] | Xi'an University of Electronic Science and technology -- Jiao Licheng: the challenge and thinking of deep learning in the era of artificial intelligence

Michael I. Jordan: Computational Thinking, inferential thinking and data science

[original] Wang Feiyue: opportunities and challenges of generative network confrontation

[original] | Microsoft Asia Research Institute Liu Tieyan: Advanced Learning frontier

Professor of Tsinghua University: Sun Fuchun: artificial intelligence and robot in the cognitive age

When Bayes meets big data

University of California, Berkeley: Ma Yi: low dimensional structure and high dimensional model (visual) data

Professor, computer department, University of California, Berkeley: Dawn song -- how AI can build a stronger defense system

Tsinghua University: Liu Zhiyuan -- knowledge representation learning and its application

Li Feifei and the power of Imagenet

Professor of Tsinghua University: Sun Fuchun -- Prospect of core technology of educational robot

From the perspective of high-end talent allocation, the results of the fourth round of the first level discipline evaluation of computer science and technology

Huang Tiejun, Changjiang Scholar and professor of Peking University: the past, present and future of artificial intelligence

© 2023