应国家防伪工程技术研究中心邀请，悉尼科技大学量子计算与智能系统（QCIS）中心主任Chengqi Zhang教授，Ivor W Tsang教授等学者，将来我系进行大数据分析与研究系列学术讲座，欢迎广大教师和研究生参加。相关安排如下：
时间：4月16日下午2点
地点：国家防伪工程技术研究中心2楼会议室
主办：国家防伪工程技术研究中心
讲座之一：Overviewof research in the Centre for Quantum Computation & Intelligent Systems
主讲人：悉尼科技大学QCIS中心主任Prof Chengqi Zhang
讲座之二：FeatureSelection for Big Data with Trillion Dimensions
主讲人：悉尼科技大学QCIS中心A/Prof Ivor W Tsang
讲座之三：FrequentPattern Mining Over Uncertain Data
主讲人：悉尼科技大学QCIS中心Dr Ling Chen
讲座之四：EfficientQuery Processing On massive spatial Data
主讲人：悉尼科技大学QCIS中心Dr Ying Zhang
讲座之五：GraphProcessing in the Era of Big Data
主讲人：悉尼科技大学QCIS中心Dr Lu Qin
讲座摘要：
1．Speaker:Prof Chengqi Zhang
Title: Overview of research in theCentre for Quantum Computation & Intelligent Systems
Abstract:
The Centre for Quantum Computation & Intelligent Systems (QCIS) was established in April 2008 as a priority investment Centre of the University of Technology, Sydney. The Centre's research focus is to develop theoretical foundations, innovative technology and practical systems that will result in next-generation enterprise intelligent information systems. Over the last six years, QCIS staff have published around 400 high quality papers in prestigious journals and conference proceedings, achieved 37 Australian Research Council grants (around AU$12.5 million), and helped to raise the UTS-IT world-wide ranking to 100-150. This talk will summarise the achievements of QCIS over last six years, and discuss the reasons behind these achievements.
Short bio: Chengqi Zhang has been a Professor of Information Technology at the University of Technology, Sydney (UTS) since December 2001. He has been the Director of the UTS Priority Investment Research Centre for Quantum Computation & Intelligent Systems (QCIS) since April 2008. He has been Alternative Dean of UTS Graduate Research School since April 2013. He has been Chairman of the Australian Computer Society’s National Committee for Artificial Intelligence since November 2005. Prof Zhang obtained his Bachelor degree from Fudan University in 1982, his Master degree from Jilin University in 1985, his PhD degree from The University of Queensland in 1991, and he followed these with a Doctor of Science (DSc – Higher Doctorate) from Deakin University in 2002 – all in Computer Science. Prof Zhang’s research interests mainly focus on Data Mining and its applications. He has published more than 200 research papers, including a number of papers in first-class international journals such as Artificial Intelligence, IEEE and ACM Transactions. He has published seven monographs and edited 16 books. He has delivered 14 keynote/invited speeches at international conferences. He has attracted 11 Australian Research Council grants. Due to his outstanding research achievements, he was awarded the 2011 NSW Science and Engineering Award in the Engineering and ICT category. Since his appointment as the Director of QCIS six years ago, Prof Zhang has led the Centre's researchers to publish around 400 high quality papers, double the national grant funding, and improve the grant ranking of UTS-IT to 100-150 world-wide, based on the Shanghai Jiao Tong University ranking. His Centre has achieved seven ARC Future Fellows in the last five ARC rounds, which is 18% of the national share from 38 Universities. Due to his leadership achievements, Prof Zhang was awarded the 2011 UTS Vice-Chancellor's research excellence awards in the Leadership category. Prof Zhang is a Fellowof the Australian Computer Society (ACS) and a Senior Member of the IEEE Computer Society (IEEE). He has been serving as an Associate Editor for three international journals, including IEEE Transactions on Knowledge and Data Engineering from 2005 to 2008; and he served as General Chair, PC Chair, or Organising Chair for five international Conferences, including ICDM 2010 and WI/IAT 2008. He is also General Co-Chair of KDD 2015 in Sydney and the Local Arrangements Chair of IJCAI-2017 in Melbourne (International Joint Conference on Artificial Intelligence).
|
2. Speaker: A/Prof Ivor W Tsang
Title: FeatureSelection for Big Data with Trillion Dimensions
Abstract:
The world continues to generate quintillion bytes of data daily, leading to pressing needs for new endeavours to deal with the grand challenges brought about by Big Data. Today, there is consensus between the machine learning and data mining communities that data volume presents an immediate challenge pertaining to scalability. However, when addressing volume in Big Data analytics, researchers have taken a one-sided view of volume, which is the "big instance size" factor of the data. The flip side of volume, which is the dimensionality factor of Big Data, has received much less attention. In this talk, I will attempt to fill this gap and place special focus on the relatively under-explored topic of ultra-high dimensionality. Specifically, I will first reformulate the resultant non-convex problem as a convex semi-infinite programming (SIP) problem, and then present an efficient feature generation paradigm to solve it. The proposed feature generation will achieve lower feature selection bias compared to L1-regularized methods. In addition, trillions of correlations among millions of features can be efficiently handled by the proposed feature generation framework. Comprehensive experiments on a wide range of synthetic and real-world datasets with tens of millions of data points and O(10^14) dimensions demonstrate that the proposed method achieves superb performance when compared with state-of-the-art feature selection methods, in terms of generalisation performance and training efficiency. |
Short BIO:
Ivor W Tsang is an Australian Future Fellow and Associate Professor with the Centre for Quantum Computation & Intelligent Systems (QCIS), at the University of Technology, Sydney (UTS). Before joining UTS, he was the Deputy Director of the Centre for Computational Intelligence, Nanyang Technological University, Singapore. He received his PhD degree in computer science from the Hong Kong University of Science and Technology in 2007. His research focuses on kernel methods, transfer learning, feature selection, big data analytics for data with millions of dimensions, and their applications to computer vision and pattern recognition. He has more than 100 research papers published in refereed international journals and conference proceedings, including 4 JMLR, 8 T-PAMI, 18 T-NN, 12 ICML, NIPS, UAI, AISTATS, SIGKDD, IJCAI, AAAI, ICCV, CVPR, ECCV, ACL, etc. Dr Tsang received his prestigious Australian Research Council Future Fellowship in 2013, and had previous been awarded the 2008 Natural Science Award (Class II) by the Ministry of Education, China, and the IEEE Transactions on Neural Networks Outstanding 2004 Paper Award in 2006. His research also earned him the Best Student Paper Award at CVPR'10, the Best Paper Award at ICTAI'11, and the Best Poster Honorable Mention at ACML'12. He was also conferred with the Microsoft Fellowship in 2005. |
3. Speaker: Dr Ling Chen
Title: Frequent Pattern Mining Over Uncertain Data
Abstract:
Data uncertainty isinherent in various applications. It has also posed many unique challenges tonearly all types of data mining tasks, creating a need for uncertain datamining. This talk introduces our recent research on mining frequent patternsover uncertain data. In particular, it focuses on the following two topics: 1)frequent serial episode mining over uncertain data, and 2) probabilisticfrequent pattern summarisation.
For frequent serialepisode mining, research is undertaken to devise data mining algorithms todiscover independent frequent serial episodes and dependent frequent serialepisodes. We have developed exact solutions and approximate solutions, as wellas optimisation solutions for mining the two types of frequent patterns. Ourwork was published at EDBT2013 and ICDM2013.
For probabilisticfrequent pattern summarisation, we have formally defined the problem ofprobabilistic representative frequent pattern mining, which aims to find theminimal set of patterns with sufficiently high probability to represent allother patterns. The developed exact solution and approximate solution werepublished at SDM2013 and KDD2013, respectively.
Short BIO:
Dr Ling Chen is aLecturer with the Faculty of Engineering and Information Technology (FEIT),University of Technology, Sydney. She received her PhD in 2008 in ComputerEngineering, from Nanyang Technological University, Singapore. She was aPostdoctoral Research Fellow with the L3S Research Centre, Leibiniz UniversityHannover, Germany from 2007 to 2009. Ling's main research interests includedata mining, machine learning and social media. Ling is currently a core memberof the Centre for Quantum Computation & Intelligent Systems (QCIS) at UTS.
4. Speaker: Dr Ying Zhang
Title: Efficient Query Processing Onmassive spatial Data
Abstract:
Withthe rapid development of positioning technologies and the boosting deploymentof inexpensive location-aware sensors, a large volume of spatial data has beenrapidly increasing. The value locked up in the overwhelming amounts of spatialdata presents extraordinary and unprecedented opportunities to discover andshare new knowledge in many critical applications, such as location basedservice (LBS), national security and defence, marketing, traffic management, healthcare and environmental monitoring. This talk will introduce important researchwork on massive spatial data processing with special focus on uncertain spatialdata processing and spatial keyword search, including some of our recentcontributions in these fields. I will also discuss our recent research work onstreaming enriched geo-spatial data.
Short BIO:
Dr Ying Zhang is asenior lecturer and ARC DECRA Research Fellow at the University of Technology,Sydney (UTS) . He received his BSc and MSc degrees in Computer Science fromPeking University, and his PhD in Computer Science from the University of NewSouth Wales. His research interests include query processing on data streams,uncertain data, spatial data and graphs. He has published 20+ papers in themost prestigious international journals and conference proceedings, such asTODS, VLDBJ, TKDE, SIGMOD, SIGIR, VLDB and ICDE. He also serves as a PC memberfor many international database conferences such as PVLDB and ICDE. He wasawarded an Australian Research Council Australian Postdoctoral Fellowship (ARCAPD, 2014-2016) and an Australian Research Council Discovery Early CareerResearcher Award (ARC DECRA, 2011-2013).
5. Speaker: Dr Lu Qin
Title: GraphProcessing in the Era of Big Data
Abstract:
Withthe emergence and rapid proliferation of applications that deal with biggraphs, such as web graphs (Google, Yahoo), social networks (Facebook,Twitter), e-commerce networks (Amazon, Ebay), and road networks, graphprocessing has become increasingly prevalent and important in recent years.However, in the era of big data, the explosion and profusion of available graphdata in a wide range of application domains raises new challenges and providesnew opportunities in graph processing.
Inthis talk, I will present my group’s recent progress in graph processing interms of new graph query semantics and new computing paradigms. For new graphquery semantics, I will introduce our work on structural keyword search,shortest path computation, graph matching, and community detection. For newcomputing paradigms, I will present our research on multi-core graph processingtechniques, external graph processing techniques, and distributed graphprocessing techniques (MapReduce, BSP) in the cloud. Finally, I will discusspotential future research directions for graph processing.
Short Bio:
InJuly of 2006, Dr Lu Qin received his bachelor degree from the Department ofComputer Science and Technology in Renmin University, China (RUC). He had beenthe team leader of RUC for the ACM International Collegiate Programming Contest(ICPC). His team won three gold medals and two silver medals in the ACM/ICPCregional contests and advanced to the world finals twice. In August 2010, Dr LuQin received his PhD degree from the Department of Systems Engineering andEngineering Management (SEEM) in the Chinese University of Hong Kong (CUHK). Hewas a postdoctoral research fellow inCUHK from August 2010 to August 2013. Following this, Dr Qin joined theUniversity of Technology, Sydney (UTS) as a core member in the Centre forQuantum Computation & Intelligent Systems (QCIS). He is also an adjunctlecturer at the University of New South Wales (UNSW), Australia.
DrQin’s research interests include algorithm design and analysis for new big dataproblems, big graph processing in the cloud, and big graph searching andmining. He has been very productive in databases. In the last five years, DrQin has published more than 30 top conference/journal papers, including fourSIGMOD papers, five PVLDB papers, and five ICDE papers in the top-threedatabase conferences, and five VLDB journal papers, one Algorithmica paper, andone TKDE paper in top-ranked database and algorithm journals. His book entitled“Keyword Search in Databases” is the first monograph on this field of research.Dr Qin has served as a program committee member for many top database and datamining conferences. He has received several research grants from the Australiagovernment, the Hong Kong government, and from UTS.
DrQin has extensive experience in student supervision. Over the last four years,he has supervised/co-supervised five PhD students, two of whom havesuccessfully received their PhD degrees, and three of whom have continuouslypublished their work in top-ranked conferences and journals under thesupervision of Dr Qin.
