Abstract: Automated machine learning (AutoML) on graphs, which combines the strength of graph machine learning and AutoML, is gaining attention from the research community. This talk will first overview graph machine learning and AutoML on graphs. Then, recent advances, including efficient neural architecture search for self-attention representation, hyper-parameter optimization on large-scale graphs, and increasing explainability in AutoML on graphs, will be discussed. We will also introduce AutoGL, the first dedicated framework and open-source library for AutoML on graphs, which is expected to facilitate the research and application in the community. Last but not least, our insights on future research directions will be shared with the audience.

BIO: Automated machine learning (AutoML) on graphs, which combines the strength of graph machine learning and AutoML, is gaining attention from the research community. This talk will first overview graph machine learning and AutoML on graphs. Then, recent advances, including efficient neural architecture search for self-attention representation, hyper-parameter optimization on large-scale graphs, and increasing explainability in AutoML on graphs, will be discussed. We will also introduce AutoGL, the first dedicated framework and open-source library for AutoML on graphs, which is expected to facilitate the research and application in the community. Last but not least, our insights on future research directions will be shared with the audience.
He served as the Editor-in-Chief for the IEEE Transactions on Multimedia (T-MM) from January 1, 2017 to December 31, 2019. He has been serving as the chair of the steering committee for IEEE T-MM and Vice EiC for IEEE Transactions on Circuits and Systems for Video Technology (TCSVT) since January 1, 2020. He served as co-Chair for ACM MM 2018 and co-TPC-Chair for ACM MM 2014. His current research interests are in the areas of multimodal big data and intelligence, and multimedia networking including edge computing. He received nine Best Paper Awards, including ACM MM 2012. He is an IEEE Fellow, AAAS Fellow, and SPIE Fellow.

Abstract: The public blockchain, such as Bitcoin, Ethereum, and EOS, is easy to be used for money laundering, ransom payment, dissemination of illegal content, and other criminal activities due to its decentralization and anonymity, which may cause serious harm to financial order and social stability. Thus, it is urgent to study the corresponding supervision methodologies, by the comprehensive analysis of transactions and data flow, the effective realization of network monitoring and transaction traceability of public blockchain. On the other hand, the characteristics of the public blockchain, such as peer-to-peer communication and anonymous identity, are conducive to covert data transmission. The blockchain transactions can be considered as a natural carrier for the construction of covert channels. Also, the blockchain-based covert communication also faces many technical challenges. This report mainly introduces the technical characteristics of the public blockchain, as well as the supervision methodology for the public blockchain and the corresponding covert communication technology.

BIO: Liehuang Zhu, doctor, professor and doctoral supervisor, was selected into the National Special Support Program for High-level Talents. Currently, he is the director of Institute of Advanced Network and Data Security, School of Cyberspace Science and Techonoly, Beijing Institute of Technology. He also serves as secretary-general of the Blockchain Special Committee of the Chinese Computer Society, member of the Cyberspace Security Teaching Steering Committee of the Ministry of Education, and chairman of the Intelligent Information Network Professional Committee of the Chinese Artificial Intelligence Society. He also serves as the editorial board member of IEEE IoT-J, IEEE TVT, IEEE Network and other international journals. In recent years, he has presided over more than 20 national, provincial and ministerial level projects such as key R & D projects of the Ministry of Science and Technology, key projects of the National Natural Science Foundation of China, and key projects of National Defense Basic Scientific Research, published more than 150 high-level academic papers, and won 3 provincial and ministerial level science and technology awards.

Abstract: This talk presents a global trajectory of signal processing for analysis and representation emerging within decades. It begins with time-frequency decompositions, wavelet orthogonal bases, multiscale geometric analysis, approximations and sparsity from orthonormal bases to redundant dictionaries of waveforms, compressive sensing and dictionary learning as well. The striking efforts in deep learning would also be addressed for a comprehensive perspective. Our recent works exploring structured sparsity and deep convolutional networks for scalable and compact representation of high-dimensional signals will be described as some examples. By rethinking the cutting-edge signal processing techniques in alignment with interdisciplinary inspirations, it is appealing to make the scientific community move forward and avoid dying by starvation in a progressively narrower and specialized domain. If possible, we might present the trajectory from graph signal processing towards graph neural networks. Specifically, we overview existing graph network methodology and elaborate the underlying theoretical foundation. Furthermore, our relevant theoretical and technical efforts have been provided to inspire possible expectations.

BIO: Hong-Kai XIONG is a Distinguished Professor in both Dept. Electronic Engineering and Dept. Computer Science and Engineering, Shanghai Jiao Tong University (SJTU). His research interests mainly focus on multimedia signal processing, image and video coding, multimedia communication and networking, computer vision, biomedical informatics, machine learning. He published over 280 refereed journal and conference papers. Currently, he is the associate editor of IEEE Transactions on Circuits and Systems for Video Technology (TCSVT).
Since he received his Ph.D. degree from SJTU in 2003, he has been with Department of Electronic Engineering in SJTU. During 2007-2008, he was a research scholar in the Department of Electrical and Computer Engineering of Carnegie Mellon University (CMU). From 2011 to 2012, he was a scientist with the Department of Biomedical Informatics at the University of California, San Diego (UCSD). He was the recipient of the Science and Technology Innovative Leader Talent in Ten Thousand Talents Program in 2017, Yangtze River Scholar Distinguished Professor Award from Ministry of Education, China in 2016, National Science Fund for Distinguished Young Scholar Award from Natural Science Foundation of China (NSFC) in 2014, Shanghai Youth Science and Technology Distinguished Achievement Award in 2018, Shanghai Academic Research Leader Talent Award in 2017, Shanghai Youth Science and Technology Talent Award in 2014, Shanghai Shu Guang Scholar Award in 2013, 2017 Baosteel Excellent Faculty Award, New Century Excellent Talent Award from Ministry of Education of China in 2009, SJTU SMC-A Excellent Young Faculty Awards in 2013 and 2010. He has been granted 2017 and 2011 First Prizes of the Shanghai Technology Innovation Award for research achievements on Network-oriented Video Processing and Dissemination.

Abstract: Online social media contains rich emotional information of online users, which has become the best resource for countries and enterprises to understand the emotions and opinions of netizens. Web text sentiment analysis technology has become a research hotspot in many disciplines. At the same time, emotional cognitive computing is also the highest form and frontier subject of artificial intelligence technology. The report will first give the definition of sentiment computing, and then some confusions and thoughts that he has encountered over the years will be discussed. The speaker will focus on analyzing the development trends and challenges in the field of online text sentiment computing, and then propose a binary computing model for fine-grained sentiment analysis-oriented online text sentiment cognition, and introduce the method of fine-grained sentiment analysis based on knowledge guidance. Finally, the speaker will show the application prospects of the binary emotion computing model.

BIO: Huang Yongfeng is a doctor, professor, doctoral supervisor, outstanding teacher of the first national network security, deputy director of the Institute of Information Cognition and Intelligent Systems, and a special lecture professor in Minjiang. He is also an IEEE Senior Member, Vice Chairman of ACM SIGWEB China Branch, Vice Chairman of the Artificial Intelligence and Education Committee of the China High-Tech Association, and a member of the China Information Hiding and Multimedia Security Professional Committee. For many years, he has been engaged in the research and teaching of Internet and its information security theory and technology. He has taken charge of the key projects and general projects of the National Natural Science Foundation of China, National Program on Key Basic.
Research Project of China (973 Program), National High-tech R&D Program of China (863 Program), and more than forty scientific research projects such as key research and development projects. He has served on the editorial board of many internationally renowned journals, as well as the program committee or chairman of important international conferences.
He has published more than three hundred academic papers in famous domestic and foreign journals such as Science (Eletters), Nature SR, IEEE Tran. ACM Tran., Science in China, and important international conferences such as AAAI and ACL (including more than one hundred papers included in SCI). He has published six monographs, two translations, and two textbooks. He has applied for more than twenty invention patents (eleven of them have been authorized and two technology transfers). He has won two first prizes, three second prizes of provincial and ministerial scientific and technological achievement awards, one first prize of excellent textbooks, and two teaching achievement awards of Tsinghua University.