Speakers -2022 4th International Conference on Information Science, Electrical and Automation Engineering（ISEAE 2022）

Keynote Speakers

杨凯.jpg

Prof. Kai Yang, Huazhong University of Science and Technology, China

杨凯教授，华中科技大学

Title: Coming Soon......

Abstract: Coming Soon......

Experience: Visiting scholar at the University of Leuven in Belgium and visiting professor at the University of Changwon in Korea. In recent years, he has presided over four National Natural Science Fund projects, two major national science and technology projects, one national support plan, one 863 Project, one provincial fund, one international cooperation project, and more than 40 other school-enterprise cooperation projects. He has published more than 200 papers of the first author in the important academic journals at home and abroad, among which 170 papers are included in Sci/ei. There are 25 patents for invention and 38 patents for utility model.

ISEAE 2021

Prof. Xiaoqiang Yan, University of Science & Technology Beijing

闫晓强教授，北京科技大学

Title: Vibration control technology and industrial application of modern large rolling mill
Abstract: Rolling mill vibration is an urgent technical problem in the field of rolling in the world in the past 60 years. When rolling high strength thin strip steel, the rolling mill often has strong vibration phenomenon, which leads to defects such as vibration marks on the surface of strip and roll. It seriously affects the surface quality of strip steel and can not meet the needs of high-end users. At the same time, it also threatens the safety production of rolling mill and reduces the service life of parts, which has become the bottleneck of high-quality thin strip production in iron and steel enterprises.
After 13 years of research on the vibration of 23 rolling mills in China, the research group has expand from the traditional research methods to the vibration energy, from the single parameter research to the multi parameter coupling effect research, from the influence factors of rolling mill itself to the whole flow field of casting and rolling, and found the "ghost" of inducing rolling mill vibration. The theory system of multi state coupling vibration in electromechanical and hydraulic field of rolling mill is established, the heredity of vibration gene in casting and rolling process is revealed, the coupling vibration decoupling vibration suppressor of rolling mill is invented, and the cooperative control technology of continuous casting and rolling mill is developed, so that the rolling mill vibration is significantly suppressed and the enterprise obtains significant economic benefits.
In 2018, the appraisal expert group of China metal society and China Heavy Machinery Industry Association in 2020 agreed that the achievement has reached the international leading level.
In recent three years, rolling mill vibration research has won three first prizes of provincial and ministerial level scientific and technological progress, and has become a key promotion project of China Society of metals.

Prof. Jie Zhu, Shanghai Jiao Tong University
朱杰教授，上海交通大学

Title: The Efficient and Scalable Artificial Intelligence in Healthcare

Abstract: As the pace of life and work accelerates, it is important to monitor the health of individuals on a daily basis. However, few of us can get access to such treatment because of the expensive cost and the lack of time. Now, artificial intelligence enables people to make use of the professional treatments which could only be provided by specialists in the past. We have done a lot of works in the healthcare fields. For example, we established an emotional disorder diagnosis system to learn and judge people's mental state through people's voice information by a deep learning based method. For autism spectrum disorder (ASD), we built an intelligent diagnosis system based on distinguishing small differences in fine-grained action understanding and recognition in videos. Clinical trials have proven that the proposed diagnosis systems can obtain a high-confidence diagnosis results, which have certain advantages among similar researches. Having said that, the superior performance comes from the deeper and wider neural networks, which leads to high computational cost and high memory footprint. As a result, such neural networks are hard to be deployed on people’ s daily life, i.e., wearable devices. In order to alleviate this issue, we have put forward to several structured network pruning methods and knowledge distillation based methods. Experimental results show that the proposed methods can greatly compress the neural networks with negligible accuracy loss.

Prof. Feng Zhao, University of Science and Technology of China
赵峰教授，中国科学技术大学
Title: Machine Learning Using Biometrics – An Introduction

Abstract: Machine learning techniques have been widely used in many fields, such as fingerprint recognition, face recognition, speech recognition, natural language processing, and so on. In this talk, I will give an introduction to biometrics-based machine learning, including the concepts, seven key steps, applications and benefits.