Abstract: We propose a simple yet effective multiple kernel clustering algorithm, termed simple multiple kernel k-means (SimpleMKKM). It extends the widely used supervised kernel alignment criterion to multi-kernel clustering. Our criterion is given by an intractable minimization-maximization problem in the kernel coefficient and clustering partition matrix. To optimize it, we re-formulate the problem as a smooth minimization one, which can be solved efficiently using a reduced gradient descent algorithm. We theoretically analyze the performance of SimpleMKKM in terms of its clustering generalization error. Comprehensive experiments on 11 benchmark datasets demonstrate that SimpleMKKM outperforms state of the art multi-kernel clustering alternatives.
Biography：Xinwang Liu received his PhD degree from National University of Defense Technology (NUDT), China. He is now Professor at School of Computer, NUDT. His current research interests include kernel learning, multi-view clustering and unsupervised feature learning. Dr. Liu has published 60+ peer-reviewed papers, including those in highly regarded journals and conferences such as IEEE T-PAMI, IEEE T-KDE, IEEE T-IP, IEEE T-NNLS, IEEE T-MM, IEEE T-IFS, NeurIPS, CVPR, ICCV, AAAI, IJCAI, etc. He is now one of the associated editor of Information Fusion Journal. More information can be found at https://xinwangliu.github.io/
Most real systems consist of a large number of interacting, multi-typed components, which can be modeled as heterogeneous information networks. Recently, heterogeneous information network has attracted more and more attentions, since it contains rich structure and semantic information through fusing heterogeneous information. Meanwhile, with the surge of deep learning, network embedding has shown its powerful ability to learn the feature representation of nodes for downstream tasks. However, contemporary network embeddings focus on homogeneous network. In this talk, I will introduce our recent work on heterogeneous information network embedding. Moreover, some newest work on dynamic heterogeneous network embedding will also be discussed.
Chuan Shi is the professor in School of Computer Sciences of Beijing University of Posts and Telecommunications, deputy director of Beijing Key Lab of Intelligent Telecommunication Software and Multimedia. The main research interests include data mining, machine learning, artificial intelligence and big data analysis. He has published more than 100 refereed papers, including top journals and conferences in data mining, such as IEEE TKDE, ACM TIST, KDD, AAAI, IJCAI, and WWW. And in the meanwhile, his first monograph about heterogeneous information networks has been published by Springer. He has been honored as the best paper award in ADMA 2011 and ADMA 2018, and has guided students to the world champion in the IJCAI Contest 2015, the premier international data mining competition. He is also the recipient of “the Youth Talent Plan” and “the Pioneer of Teacher's Ethics” in Beijing.
A knowledge base is a large repository that typically stores information about real world entities. Several efforts have been made to develop knowledge bases in general and specific domains such as DBpedia, YAGO, LinkedGeoData, and Wikidata. These knowledge bases contain millions of facts about entities. However, they are far from complete and mandate continuous enrichment and curation. In this talk, we present three methods to enrich a knowledge base. The first is a Knowledge Bases Alignment method that finds entities in two knowledge bases which represent the same real-world entity, and then integrates these knowledge bases based on the aligned entities. The second is a Relation Extraction method that extracts entities and their relationships from sentences of a corpus and map them to an existing knowledge base. The third is a Description Generation method that generates a sentence to describe a target entity from its properties in a knowledge base.
About the speaker:
Dr. Jianzhong Qi is a senior lecturer in the School of Computing and Information Systems at The University of Melbourne. He obtained his PhD degree from The University of Melbourne in 2014. He has been an intern at Microsoft Redmond in 2014 and a visiting scholar at Northwestern University in 2017. Jianzhong Qi publishes in leading venues in database management and artificial intelligence such as TPAMI, TODS, VLDBJ, ICML, NeurIPS, and PVLDB. He has won the “Best Vision Paper” award in ACM SIGSPATIAL 2017. He is the PC Co-Chair for the Australasian Database Conference 2020. His research interests include spatio-temporal databases and artificial intelligence.
Abstract: Civil engineering infrastructure is generally the most expensive national investment and asset of any country. It requires much longer service life compared to other commercial products because civil structures are costly to replace once they are constructed. All structures, such as bridges, wind energy plants, oil rigs, high-rise buildings, require regular maintenance because they are subject to various internal and external factors which could cause malfunction. Moreover, each structure is unique in terms of materials, design, and construction. To ensure the structural integrity and safety, civil structures must be equipped with Structural Health Monitoring (SHM) aiming to develop automated systems for the continuous monitoring, inspection, and damage detection of structures with minimum labour involvement to reduce cost and risk to humans. In this talk, Jianxin Li will overview the mainstream deep learning techniques used for structural health monitoring in different environments, present their existing challenges and limitations, and then discuss the research opportunities of this area from a new perspective in terms of knowledge graph. This talk will be favour of all attendees who may have interest about interdisciplinary research of data science, AI, and structural engineering.
Dr Jianxin Li is an A/Professor of Data Science, and Director of Smart Networks Lab, in the School of IT, Deakin University. His research interests include social computing, query processing and optimization, and big data analytics. He has published 90 high quality research papers in top international conferences and journals, including PVLDB, IEEE ICDE, SIGKDD, ACM WWW, AAAI, IJCAI, IEEE ICDM, EDBT, ACM CIKM, IEEE TKDE, The VLDB Journal, IEEE TII, IEEE Internet of Things, and WWW Journal. Jianxin is serving many academic committees. He is the associate editor for Information Systems, IEEE Signal Processing Letters, and the editor-in-chief of the International Journal of Knowledge Science & Engineering. He is the guest editor for many other international journals, such as IEEE Transactions on Industrial Informatics, Neurocomputing, Computational Intelligence, IET Intelligent Transport Systems, Complexity, Data Science and Engineering. Jianxin also serves for many international conference committees, such as the General Co-chair at the 18th IEEE International Symposium on Parallel and Distributed Processing with Applications 2020, the PC Co-chair at the 15th International Conference on Advanced Data Mining and Applications 2019, PC members for world leading conferences like PVLDB, ICDE, AAAI, IJCAI, and invited reviewers for IEEE TKDE, ACM TKDD, WWW Journal and VLDB Journal. His research has been funded by Australia Research Council Discovery Project, Linkage Project, Western Australia State Government, and several industry grants..