Biography: 

Jinping Ao (Ph.D.), born in 1967, a native of Jiangxi Province, a selected candidate of the National High-level Talent Program, and the Chief Scientist of the National 13th Five-Year Plan for Key Research and Development (N13th Five-Year Plan), graduated from the Department of Physics of Wuhan University with a Bachelor's Degree of Science in 1989, and obtained a Master's Degree of Semiconductor Physics and Devices from the Thirteenth Research Institute of the Ministry of Electronics Industry (now the Thirteenth Research Institute of the China Electronics Technology Corporation (CETIC)) in 1992, and a D. degree in Microelectronics and Solid State Electronics from Jilin University in 2000. He was a senior engineer and deputy director of GaAs ultra-high-speed integrated circuits research lab in the 13th Research Institute of the Ministry of Electronics Industry, engaged in GaAs high-speed electronic devices, integrated circuits and optoelectronic integrated circuits. He has presided over many national projects such as 863 program, pre-research and national research program, etc. In 2001, he went to the University of Tokushima, Japan, as an associate professor and doctoral supervisor, engaging in the research of optoelectronic devices and electronic devices based on wide bandwidth semiconductors. He has led or participated in a number of projects such as Japan Aid for Scientific Research, JST, SCOPE and NEDO. He has many years of cooperation with famous Japanese companies such as Toyota, Sumitomo Electric, Nichia Chemical, etc. In 2016, he was selected for the National High-level Talent Program, and is a professor and doctoral supervisor at Xi'an University of Electronic Science and Technology. As a project leader, he has completed the project of “GaN-based new power electronic device key technology” in the National 13th Five-Year Plan of Key Research and Development Program “Strategic Advanced Electronic Materials”, and has joined the School of Internet of Things Engineering of Jiangnan University as a professor and Ph. He has joined the School of Internet of Things Engineering of Jiangnan University as a professor and doctoral supervisor since 2022. He has published more than 300 papers in international academic journals and conferences, and holds more than 20 patents. He was awarded the Third Prize of Science and Technology Progress Award of Ministry of Electronics Industry, and the First Prize of Shaanxi Provincial Science and Technology Award.

Biography: 

Prof. Tao Liu received his Ph.D. degree in Intelligent Mechanics from Kochi University of Technology, Japan, in September 2006, and has since stayed at the Department of Mechanical Engineering, Kochi University of Technology, Japan, where he has been a postdoctoral researcher, an assistant professor, a special lecturer, and a visiting professor. Since January 2014, he has been a professor and a doctoral director of the Institute of Mechatronics Control Engineering, Department of Mechanical Engineering, Zhejiang University, and in 2015, he was promoted to Senior Member of IEEE, and in December 2017, he became the director of the Institute of Micro-Nano Technology and Precision Engineering, Zhejiang University. For the first time in the world, he has developed a wearable sensor system applied to the analysis of human body dynamics, solving the international technical problem that the quantitative analysis of human joint and muscle forces can only rely on high-speed camera motion capture and fixed force plate systems. He has led and participated in 13 projects of Japan Society for the Promotion of Science, Japan Society for Science and National Natural Science Foundation of China. The results of international invention patents have been productized and applied in research institutes in the U.S.A., Japan, and Canada, and won the 2010 Japan Society of Mechanical Engineers Reward Award (20 places/year, 1 place/discipline). He has published more than 100 papers and served as Associate Editor of international journals IEEE Robotics and Automation Letters and IEEE Access, Associate Editor of international conferences IEEE AIM 2009-2021, IROS 2019-2021, URAI 2010-2014, JSME Welfare Symposium, Japan 2009, and JSME D&D Annual Meeting 2011, Japan, and other conference program committee member. He serves as an expert for reviewing the FCT project of the European Union Science Foundation, a member of the National Working Group for Standardization of Special Operation Robotics, and a long-term reviewer of more than 20 journals of IEEE, IOP, and IMEKO, among other societies.

Biography: 

Researcher of the Hundred Talents Program of Zhejiang University, Ph.D. He graduated from the Department of Electronic Engineering of Tsinghua University in 2010, and Ph.D. graduated from the Department of Computer Science and Engineering of the Hong Kong University of Science and Technology in 2015, and has been a post-doctoral fellow at the University of Toronto, Canada, and a researcher at Wuhan University. He is mainly engaged in the research of intelligent IoT security and has published more than 100 papers in international authoritative journals and conferences in the fields of computer network and information security. He has won the first prize of Zhejiang Science and Technology Progress Award, the Natural Science Award of the Chinese Institute of Electronics, and has been selected for the “China Association for Science and Technology Young Talent Support Project” and the “Green and Orange Prize Most Potential Award” of Dharma Institute. He is a program committee member of ACM CCS, USENIX Security, NDSS, IEEE INFOCOM and other international conferences. He serves as an editorial board member of IEEE TIFS and other international journals.

Title: Backdoor Attack Detection for Deep Learning Models

Abstract: 

Deep learning models are widely applied in critical domains such as facial recognition and autonomous driving, where backdoor attacks pose a significant risk. Backdoor attacks are highly stealthy and come in various forms, making it challenging for existing detection methods to identify continuously evolving attack techniques. This report introduces a backdoor detection method based on latent features of models. Fundamentally, backdoored models exhibit differences in latent features compared to benign models, but these differences may be deliberately disguised by attackers and are thus difficult to detect. To address this problem, we propose a novel latent feature representation method that characterizes the dominance distribution among classes. This distribution can accurately distinguish backdoored models from normal ones and is resistant to tampering by attackers. Experimental results demonstrate the effectiveness of the proposed detection method against 14 typical backdoor attacks and over 10,000 models.

Biography: 

Ming Yang, born in December 1990, is a native of Yanhe, Guizhou. He is mainly engaged in the research of mechanical motion control and testing, machine vision inspection, vibration sensor calibration and application, etc. He is the associate editor of IEEE Sensors Journal (IF = 4.3), IEEE Senior Member of Chinese Academy of Sciences (CAS), and he has published more than 20 SCI papers of CAS region 2 and above, and he has presided over 3 vertical projects at national level, and he has authorized nearly 30 patents of Chinese and American inventions. He has authorized nearly 30 patents in China and the United States, won the first prize of the Science and Technology Progress Award of the Chinese Society of Metrology and Testing, and participated in the editing of one national measurement technical specification.

Biography: 

Bo Dong, male, is a Distinguished Professor at the School of Integrated Circuits and Optoelectronic Chips, Shenzhen University of Technology. He was selected for Shenzhen Overseas High-level Overseas Educated Talents Peacock Program B. He was selected for the Technical Excellence of Pioneer Action Hundred Program of the Chinese Academy of Sciences. He has been selected as one of the top 2% top scientists in the world by Stanford University and Elsevier. He is an expert in the evaluation of the National Science and Technology Prize as well as the National Key Programs. D. degree in Optics from Nankai University and Ph.D. degree in Electrical and Computer Engineering from National University of Singapore in June 2008 and December 2015, respectively. He was a postdoctoral fellow at Laurier University in Canada from 2008-2009, and worked as a scientist at A*STAR Research Institute of Infocommunications in Singapore from 2009-2017. He has led and participated in a number of collaborative projects with NRSERC (Canada), NRF (Singapore), 863 (China), NSF (China), and Airbus (Europe), ST (Singapore), SMRT (Singapore), etc. He is also a scientist at A*STAR Institute of Infocomm Research (Singapore). He has achieved a series of innovative results in photonic integrated devices, SAW/BAW chips, optoelectronic detection, etc. He has published more than 140 academic papers, applied for more than 30 patents and authorized more than 10. The results of the projects done were awarded the 2013 ASEAN Outstanding Engineering Achievement Award and Singapore Outstanding Engineering Achievement Award. He has been invited to give invited presentations at nearly 20 international conferences, and served as session chair and technical committee member of nearly 20 international conferences. He has served as a subject editor of Applied Optics, a prestigious optoelectronics journal of OSA, and as an editorial board member of Structural Monitoring and Maintenance, a journal of KAIST, Korea. He is a senior member of Optica and IEEE.

Title: Tactile Sensor Based on Flexible Thin Film Bulk Acoustic Resonator (FBAR)

Abstract: 

Flexible thin film based tactile sensors, also known as "electronic skin", they can sense parameters such as contact force, temperature, and humidity with the external environment, and are the most important sensors for intelligent robots to perceive the environment. However, currently most flexible tactile sensors are limited by power supply, signal transmission, and other factors, which restrict their widespread application. Thin film bulk acoustic resonator (FBAR) is a miniature radio frequency device based on piezoelectric thin film materials, which has the advantages of small size, light weight, high reliability, wireless transmission, etc. It has a wide range of application value in radio frequency communication and sensing fields. Passive wireless flexible tactile sensors based on FBAR can solve the key problems of power supply and signal transmission for flexible tactile sensors. This report mainly introduces some research achievements of our research group in flexible FBAR tactile sensors in recent years: flexible FBAR tactile sensors based on PVDF, flexible FBAR tactile sensors based on Pb1.2 (Zr0.52Ti0.48) O3, and FBAR humidity sensors based on PETA/Pb1.2 (Zr0.52Ti0.48) O3. Research on the Tactile Sensor of FBAR Resonator