- Keynote Speakers
Tongji University, China
Jiang Changjun, expert in computer science, academician of Chinese Academy of Engineering, vice chairman of Shanghai Association for Science and Technology, chair professor of Tongji University, honorary professor of Brunel University London, Fellow of the British Institute of Engineering Technology, winner of The National Science Fund for Distinguished Young Scholars, Chief Scientist of the National 973 Program, National Outstanding Scientific and Technological Worker, Winner of the National Innovation Award. He has been continuously devoted to the research of network computing. He created the behavior theory of concurrent systems, established the formal method of streaming decomposition, unraveled the technical problems of system concurrency decoupling, and provided independent innovative technical solutions for the high concurrency of large-scale streaming computing. He invented the concurrent scheduling technology for network resource management and optimization, overcame the technical problems of uncertainty and timeliness of concurrent computing resource allocation, developed a network concurrent computing platform, ensured the high-efficiency engineering application of the transaction risk control system, and effectively promoted the rapid development in large-scale streaming computing in China. He has proposed the behavior authentication method for online transaction risk prevention and control, and established a hierarchical diagnosis and treatment and hierarchical risk control mechanism, which breaks through the limitations of identity authentication and changes the traditional model of transaction risk control. He presided over the establishment of the architecture of China's first online transaction risk prevention and control system, standards, making pioneering contributions to make China becoming an international "leader" in this field. His research works have been positively evaluated and cited many times by well-known experts such as academicians from the United States, the United Kingdom, Germany, Sweden, India and other countries. He has been granted 106 invention patents in China, the United States, Germany and other countries, and 22 international PCTs. He published more than 300 papers (including 82 papers in ACM/IEEE Transactions), and 5 monographs in Chinese and English; He had one second prize of the National Technology Invention Award (Rank 1), two second prizes of National Science and Technology Progress Award (both ranked 1), and HO PAN CHING YI Award (independent), etc.
Title: Cabin Computing
Abstract: With the rapid development of IT innovation and the in-depth development of digitalization, networking and intelligence, the emergence of new applications and the booming of new business forms put forward new requirements for the agile construction and continuous operation and maintenance of business systems. In this talk, the evolution of network computing will be introduced, including grid computing, cloud computing, edge computing, multi-cloud and inter-cloud computing; Secondly, a new computing mode, cabin computing, is highlighted. It is accessed through the network, and it is a computing integration environment for cross domain resource allocation and collaboration in the full life cycle of IT tasks. The core is "cabin dedicated mobility, resource cross domain elasticity and system operation and maintenance autonomy". Finally, the computation network system based on cabin computing is discussed.
University of Alberta, Canada
Witold Pedrycz (IEEE Life Fellow) is Professor in the Department of Electrical and Computer Engineering, University of Alberta, Edmonton, Canada. He is also with the Systems Research Institute of the Polish Academy of Sciences, Warsaw, Poland. Dr. Pedrycz is a foreign member of the Polish Academy of Sciences and a Fellow of the Royal Society of Canada. He is a recipient of several awards including Norbert Wiener award from the IEEE Systems, Man, and Cybernetics Society, IEEE Canada Computer Engineering Medal, a Cajastur Prize for Soft Computing from the European Centre for Soft Computing, a Killam Prize, a Fuzzy Pioneer Award from the IEEE Computational Intelligence Society, and 2019 Meritorious Service Award from the IEEE Systems Man and Cybernetics Society.
His main research directions involve Computational Intelligence, Granular Computing, and Machine Learning, among others.
Title: A Society-Oriented Environment of Machine Learning
Abstract: Over the recent years, we have been witnessing spectacular achievements of Machine Learning with highly visible accomplishments encountered, in particular, in natural language processing and computer vision impacting numerous areas of human endeavours. Driven inherently by the technologically advanced learning and architectural developments, Machine Learning constructs are highly impactful coming with far reaching consequences; just to mention autonomous vehicles, health care imaging, decision-making in critical areas, among others.
We advocate that the design and analysis of ML constructs have to be carried out in a holistic manner by identifying and addressing a series of central and unavoidable societal quests. The key challenges on the list of interest concerns interpretability, energy conciseness (being also identified on the agenda of green AI), efficient quantification of quality of ML constructs, their brittleness and conceptual stability coming hand in hand with the varying levels of abstraction. These quests are highly intertwined and exhibit relationships with the technological end of ML. As such, they deserve prudent attention, in particular when a multicriterial facet of the problem is considered.
The talk elaborates on the above challenges, offers definitions and identifies the linkages among them. In the pursuit of coping with such challenges, we advocate that Granular Computing can play a pivotal role offering a conceptual environment and realizing algorithmic development. As a detailed study, we discuss the ideas of knowledge transfer showing how a thoughtful and prudently arranged knowledge reuse to support energy-aware ML computing. We discuss passive and active modes of knowledge transfer. In both modes, the essential role of information granularity is identified. In the passive approach, information granularity serves as a vehicle to quantify the credibility of the transferred knowledge. In the active approach, a new model is constructed in the target domain whereas the design is guided by the loss function, which involves granular regularization produced by the granular model transferred from the source domain. A generalized scenario of multi-source domains is discussed. Knowledge distillation leading to model compression is also studied in the context of transfer learning.
State University of New York at Buffalo, USA
Dr. Junsong Yuan is Professor and Director of Visual Computing Lab at Department of Computer Science and Engineering (CSE), State University of New York at Buffalo, USA. Before joining SUNY Buffalo, he was Associate Professor (2015-2018) and Nanyang Assistant Professor (2009-2015) at Nanyang Technological University, Singapore. He obtained his Ph.D. from Northwestern University in 2009, M.Eng. from National University of Singapore in 2005, and B.Eng. from Huazhong University of Science Technology (HUST) in 2002. He received Chancellor's Award for Excellence in Scholarship and Creative Activities from State University of New York, Nanyang Assistant Professorship from Nanyang Technological University, Outstanding EECS Ph.D. Thesis award from Northwestern University, and Best Paper Award from IEEE Trans. on Multimedia. He serves as Senior Area Editor of Journal of Visual Communication and Image Representation (JVCI), Associate Editor of IEEE Trans. on Image Processing (T-IP), IEEE Trans. on Circuits and Systems for Video Technology (T-CSVT), and Machine Vision and Applications (MVA). He also serves as General/Program Co-chair of ICME and Area Chair for CVPR, ICCV, ECCV, ACM MM, etc. He is elected Faculty Senator at Both SUNY Buffalo and NTU. His recent work are sponsored by Meta, Microsoft, Snap, Amazon, Kwai, and Oppo. He is a Fellow of IEEE and IAPR.
Title: Intelligent Hand Sensing and Applications in Metaverse
Abstract: Humans are the most intelligent beings on the planet not only because of our powerful brain but also due to the unique structure of our hands. Hands have been crucial tools for us to interact and change physical world, as well as virtual world such as metaverse. In this talk, we will discuss real-time hand sensing using optical cameras, and how it can enhance our interactions with physical world and metaverse. Towards 3D hand sensing from single 2D images, we will discuss how to leverage synthetic hand data to address high-dimensional regression problem of articulated hand pose estimation and 3D hand shape reconstruction. To improve the generalization ability of handling hands of various shapes and poses, we will also discuss invariant hand representation through disentanglement. The resulting systems can facilitate intelligent interactions in virtual and real environments using bare hands, as well as via hand object interactions.
Harbin Institute of Technology, Shenzhen, China
Yong Xu is a Professor in Harbin Institute of Technology, Shenzhen. His research interests include pattern recognition, biometric recognition, image processing, deep learning, and bioinformatics. He was the winner of the second prize of Heilongjiang Province Natural Science in 2014, and the winner of the New Century Excellent Talents of the Ministry of Education in 2008. He is also a member of Guangdong Province Talents for supporting projects, Shenzhen Pengcheng Scholars, and Shenzhen High-level Local Talents. He has won two provincial and ministerial-level scientific and technological progress awards and has hosted three national-level projects including the National Natural Science Foundation. He has published three academic monographs, two translated works and more than 130 SCI retrieved papers. In recent years, he has been consecutively selected as one of the Most Cited Chinese Researchers by Elsevier. His part-time social jobs include IEEE senior member, IEEE SMC TC on Biometrics and Applications co-chairman, editor-in-chief of the international academic journal "International Journal of Image and Graphics", multiple international academic journals (IEEE Transactions on Cybernetics, Patten Recognition, Neurocomputing, International Journal of Pattern Recognition and Artificial Intelligence, etc.), reviewer of multiple international conference papers.
Title: Weakly Supervised Object Detection
Abstract: Weakly supervised object detection (WSOD) using image-level rather than instance-level annotations (object localization and category) to train detectors has important research significance and application value. Especially compared with fully supervised object detection, WSOD does not require labeling the object location, so it has an incomparable efficiency advantage. On the other hand, it tends to suffer from the problems of misclassified and missing objects as well as inaccurate localization. This report introduces its definition, key issues, classical methods and our latest research based on adaptive instance refinement.
Tianjin University, China
Qinghua Hu received the B.S., M.S., and Ph.D. degrees from the Harbin Institute of Technology, Harbin, China, in 1999, 2002, and 2008, respectively. After that he joined Department of Computing, The Hong Kong Polytechnical University as a postdoctoral fellow. He became a full professor with Tianjin University in 2012, and now is a Chair Professor and Deputy Dean at College of Intelligence and Computing. His research interest is focused on uncertainty modeling, multi-modality learning, incremental learning and continual learning these years, funded by National Natural Science Foundation of China and The National Key Research and Development Program of China. He has published more than 300 peer-reviewed papers in IEEE TKDE, IEEE TAPMI, IEEE TNNLS, etc. He was a recipient of the best paper award of ICMLC 2015 and ICME 2021. He is an Associate Editor of the IEEE Transactions on Fuzzy Systems, ACTA AUTOMATICA SINICA, and ACTA ELECTRONICA SINICA.
Title: Trustworthy Multi-Modality Information Fusion with Lower-Quality Heterogeneous Data
Abstract: Massive heterogeneous data are collected in various domains due to the development of sensing techniques, where texts, images, videos are usually acquired for describing an object. Some typical applications include scientific exploration, medical diagnosis, social media analysis, robot and intelligent driving. Obviously, there are different and complementary information hiding the heterogeneous data, and the quality of the data dynamically fluctuates. It is a challenging and key problem to take full advantage of heterogeneous data. It is still very hot although there are a great number of researches on this topic in these years. We will review the history on multi-modality data fusion and introduce how the brain works with multi-sensing systems, and then we will introduce some new work about dynamical and trustworthy multi-modality information fusion techniques.
Xidian University, China
Jiao Licheng, a foreign member of the academia European and the Russian academy of natural sciences., IEEE/IET/CAAI/CAA/CIE/CCF Fellow He is currently the Huashan Distinguished Professor of Xidian University, Director of the Department of Computer Science and Technology, Dean of the Institute of Artificial Intelligence, Director of the Key Laboratory of Intelligent Perception and Image Understanding of the Ministry of Education, and Expert Group of Artificial Intelligence Technology Innovation of the Ministry of Education Expert, the first batch of national-level leading talents selected, the chairman of "The Belt and Road Initiative" Artificial Intelligence Innovation Alliance, the chairman of the Shaanxi Artificial Intelligence Industry Technology Innovation Strategic Alliance, and the sixth member of the Chinese Association for Artificial Intelligence - Seven-time vice chairman, selected in Elsevier's Most Cited Scholars list for eight consecutive years. The main research directions are intelligent perception and image understanding, deep learning and brain-like computing, evolutionary optimization and remote sensing interpretation. He has won the second prize of the National Natural Science Award, the Wu Wenjun Artificial Intelligence Outstanding Contribution Award, the Huo Yingdong Young Teacher Award, the title of National Model Teacher, the China Youth Science and Technology Award, and more than ten provincial and ministerial first prizes.
Title: Next Generation Deep Learning: Thinking and Some Problems
Abstract: Although deep learning technology has been greatly developed and has achieved remarkable results in many fields. However, in the process of development, there are many theoretical problems that need further research and attention. Therefore, this report focuses on discussing with you the research related to the basic theory of deep learning. First, the origin and development of deep learning are reviewed. Then, it discusses the re-understanding and re-thinking of deep learning, which leads to the basic theory that should be broken through. Then, the representation, learning and optimization theory of deep learning are discussed from three aspects: brain-inspired, physical-inspired and evolution-inspired. Finally, some thoughts on the next generation of deep learning are given.