Famous teachers 2 2022 Issue 3 "Deep Learning Security and confrontation" special responsibility editorial board recommendation paper

"Deep Learning Security and confrontation" is the third topic of "Information Security Research" in 2022. The teacher of the main responsibility editorial committee is a professor and a doctoral supervisor Tan Yu'an, the dean of the Beijing Institute of Technology. The following is recommended by Teacher Tan for everyone. Welcome everyone to read a message.

Recommended thesis

"Generation against network technology and research progress" (Author: Liang Chen, Wang Libin, Li Zhuoqun, Xue Yuan, unit: Beijing University of Information Technology Management, etc.)

Recommended reason for thesis

While becoming an important achievement and innovation highland in the information age, the Internet will also become a place for the convergence of security risks and a new source of national conflict. With the continuous development and application of deep learning technology, the security issues brought by the technology itself also need to pay attention to. Therefore, deep learning security has become a popular research field with deep learning.

This article details the origin, design ideas, scheme framework, and implementation process of the concept of genetic confrontation network technology. The structure of the article is complete, the data is true and reliable, and the academic research value in the field of deep learning security. Starting from the relevant hotspots in recent years, analyzing security issues will be analyzed in the process of generating network technology. In addition, it also enters the important application scenarios such as computer vision, medical care, finance and other fields such as images and video processing and generation, data set enhancement, time sequence prediction and other generators. According to the introduction, the shortcomings of their respective lack of each other were listed and the issues that need to be improved. In addition, the article also focuses on how to use the generating countermeasures and how to deal with sensitive data.

Responsible editorial teacher comment

Deep learning has become a strategic technology leading a new round of scientific and technological revolution and industrial change. At the same time, the safety risks faced by deep learning are also increasingly prominent. The "Generating Conducting Network Technology and Research Progress" written by the authors of Beijing University of Information Technology and other authors summarized the technology development and application of emerging technologies of generating the network against the network, and introduced the development of the formation of the formation and loss of the function of the framework. Essence From the perspective of data privacy protection, the existing generation of confrontation network technology is inadequate in satisfying the differential privacy constraints, and it is clear that the current quality of the sample set of the network generated by data privacy is limited. All in all, the article can help readers quickly understand the development and main challenges of this field in recent years, providing new research ideas for future technological innovation in the field.

The main content of the thesis

Generate adversarial networks (GAN) uses zero -harmony game theory to combine two competing neural networks to make it more clear and more discrete output. Major progress has been made in image and video processing and generating and generating in the fields of computer vision, medical care, finance and other fields.

The paper is divided into 5 chapters, which mainly introduces the basic framework, theory and implementation process of generating the network, and analyzes the current status of mainstream research in recent years. By reviewing the generating network variants and its application scenarios, it needs to be improved. In addition, this article also focuses on how to use the generating network privacy measures and processing sensitive data, and generate the future development trend of confrontation in related fields.

Full -text download link: generate -based confrontation network technology and research progress (www.sicris.cn/cn/y2022/v8/i3/235)

[Editorial Committee of responsibility in this issue · Personal profile]

Tan Yu'an, a professor and a doctoral supervisor at Beijing Institute of Technology's Network Space Security. The main research areas are deep learning confrontation and information system security. As the person in charge of the project, it undertakes more than 20 national and provincial and ministerial research projects, including key projects of the National Natural Science Foundation of China/on the project, national key research and development plan projects, etc., published more than 50 SCI papers, and has been authorized by the national authorized invention patent to patent more than 20 patents The research results were awarded 4 provincial and ministerial awards, and served as journal editorial boards such as "Information Security Research", "Information Network Security", "International Journal of Intelliggent Systems".

- END -