High Performance Computing and Cluster Technologies

Keynote & Plenary Speakers

Jane Doe
Prof. WANG Jun, City University of Hong Kong, Hong Kong
Fellow of IEEE & IAPR
王钧教授, 香港城市大学

Biography: Jun Wang is a Chair Professor of Computational Intelligence in the Department of Computer Science at City University of Hong Kong. Prior to this position, he held various academic positions at Dalian University of Technology, Case Western Reserve University, University of North Dakota, and Chinese University of Hong Kong. He also held various part-time visiting positions at US Air Force Armstrong Laboratory, RIKEN Brain Science Institute, Huazhong University of Science and Technology, Dalian University of Technology, and Shanghai Jiao Tong University as a Changjiang Chair Professor. He received a B.S. degree in electrical engineering and an M.S. degree in systems engineering from Dalian University of Technology, Dalian, China. He received his Ph.D. degree in systems engineering from Case Western Reserve University, Cleveland, Ohio, USA. His current research interests include neural networks and their applications. He published about 200 journal papers, 15 book chapters, 11 edited books, and numerous conference papers in these areas. He is the Editor-in-Chief of the IEEE Transactions on Cybernetics. He also served as an Associate Editor of the IEEE Transactions on Neural Networks (1999-2009), IEEE Transactions on Cybernetics and its predecessor (2003-2013), and IEEE Transactions on Systems, Man, and Cybernetics – Part C (2002–2005), as a member of the editorial advisory board of International Journal of Neural Systems (2006-2013), and a member of the editorial board of Neural Networks (2012-2014) as a guest editor of special issues of European Journal of Operational Research (1996), International Journal of Neural Systems (2007), Neurocomputing (2008, 2014, 2016), and International Journal of Fuzzy Systems (2010, 2011). He was an organizer of several international conferences such as the General Chair of the 13th International Conference on Neural Information Processing (2006) and the 2008 IEEE World Congress on Computational Intelligence, and a Program Chair of the IEEE International Conference on Systems, Man, and Cybernetics (2012). He has been an IEEE Computational Intelligence Society Distinguished Lecturer (2010-2012, 2014-2016). In addition, he served as President of Asia Pacific Neural Network Assembly (APNNA) in 2006 and many organizations such as IEEE Fellow Committee (2011-2012); IEEE Computational Intelligence Society Awards Committee (2008, 2012, 2014), IEEE Systems, Man, and Cybernetics Society Board of Directors (2013-2015), He is an IEEE Fellow, IAPR Fellow, and a recipient of an IEEE Transactions on Neural Networks Outstanding Paper Award and APNNA Outstanding Achievement Award in 2011, Natural Science Awards from Shanghai Municipal Government (2009) and Ministry of Education of China (2011), and Neural Networks Pioneer Award from IEEE Computational Intelligence Society (2014), among others.

Title of Speech: Collaborative Neurodynamic Optimization Approaches to Nonnegative Matrix Factorization  

Abstract: Nonnegative matrix factorization (NMF) is an advanced method for nonnegative feature extraction, with widespread applications. However, the NMF solution often entails to solve a global optimization problem with a nonconvex objective function and a nonnegativity constraint. To tackle this challenging problem, I will present a collaborative neurodynamic optimization approach by employing a population of recurrent neural networks (RNNs) at the lower level and particle swarm optimization (PSO) with wavelet mutation at the upper level. The RNNs act as search agents carrying out precise constrained local searches according to their neurodynamic equations and initial conditions. The PSO algorithm coordinates and guides the RNNs with updated initial states toward global optimal solution(s). A wavelet mutation operator is added in the optimization to enhance PSO exploration capability. Through iterative interaction and improvement of the locally best solutions of RNNs and global best positions of the whole population, the population-based neurodynamic systems is almost sure to achieve the global optimality for the NMF problem. The convergence of the group best state to the global optimal solution with probability one is proven. The experimental results substantiate the efficacy and superiority of the collaborative neurodynamic optimization approach to bound-constrained global optimization with several benchmark nonconvex functions and NMF-based clustering with benchmark datasets in comparison to the state-of-the-art algorithms. 


Jane Doe
Prof. Steven Guan, Xi'an Jiaotong-Liverpool University, China
关圣威教授, 西交利物浦大学

Biography: Steven Guan received his BSc. from Tsinghua University (1979) and M.Sc. (1987) & Ph.D. (1989) from the University of North Carolina at Chapel Hill. He is currently a Professor and the Director for Research Institute of Big Data Analytics at Xi'an Jiaotong-Liverpool University (XJTLU). He served the head of department position at XJTLU for 4.5 years, creating the department from scratch and now in shape. Before joining XJTLU, he was a tenured professor and chair in intelligent systems at Brunel University, UK.
Prof. Guan has worked in a prestigious R&D organization for several years, serving as a design engineer, project leader, and department manager. After leaving the industry, he joined the academia for three and half years. He served as deputy director for the Computing Center and the chairman for the Department of Information & Communication Technology. Later he joined the Electrical & Computer Engineering Department at National University of Singapore as an associate professor for 8 years.
Prof. Guan’s research interests include: machine learning, computational intelligence, big data analytics, mobile commerce, modeling, networking, personalization, security, and pseudorandom number generation. He has published extensively in these areas, with 130+ journal papers and 180+ book chapters or conference papers. He has chaired, delivered keynote speech for 80+ international conferences and served in 180+ international conference committees and 20+ editorial boards. There are quite a few inventions from Prof. Guan including Generalized Minimum Distance Decoding for Majority Logic Decodable Codes, Prioritized Petri Nets, Self- Modifiable Color Petri Nets, Dynamic Petri Net Model for Iterative and Interactive Distributed Multimedia Presentation, Incremental Feature Learning, Ordered Incremental Input/Output Feature Learning, Input/Output Space Partitioning for Machine Learning, Recursive Supervised Learning, Reduced Pattern Training using Pattern Distributor, Contribution Based Feature Selection, Incremental Genetic Algorithms, Incremental Multi-Objective Genetic Algorithms, Decremental Multi-objective Optimization, Multi-objective Optimization with Objective Replacement, Incremental Hyperplane Partitioning for Classification, Incremental Hyper-sphere Partitioning for Classification, Controllable Cellular Automata for Pseudorandom Number Generation, Self Programmable Cellular Automata, Configurable Cellular Automata, Layered Cellular Automata, Transformation Sequencing of Cellular Automata for Pseudorandom Number Generation, Open Communication with Self-Modifying Protocols, etc.

Title of Speech: Opportunities and Challenges in Information Communications Technology 

Abstract: This talk introduces the overall trends of Information Communications Technology (ICT) and presents an overview for opportunities and challenges in ICT. Critical issues, research problems and developments of ICT in various areas are addressed, such as green computing, Internet computing, mobile computing, and intelligent computing. Opportunities and challenges in relevant areas are also covered, for example, Internet of Things, cloud computing, big data analytics. Critical development of ICT in various aspects are proposed thereafter. Finally, the challenges faced by the higher education sector are also discussed. 

Jane Doe
Prof. Jun Xu, Institute of Computing Technology, Chinese Academy of Sciences, China
徐君教授, 中国科学院大学

Biography: Jun Xu is a professor at Institute of Computing Technology, Chinese Academy of Sciences (ICT-CAS). He received his PhD in Computer Science from the Nankai University, Tianjin, China. Before joining ICT-CAS, he was a senior researcher at Noah’s Ark Lab, Huawei Technologies at Hong Kong and an associate researcher at Microsoft Research Asian. His research interests include learning to rank for information retrieval, semantic matching in search, and text data analysis. Some of his work has published on top conferences and journals and received over 2000 citations (according to Google Scholar). He is also the author of a book: “Semantic Matching in Search” (NOW publishers). On product contributions, he has transferred several technologies to Microsoft Bing Search, Microsoft SharePoint Search, and Huawei GTS Search. On professional services, he has organized SIGIR 2014 workshop on Sematic Matching in Information Retrieval, served as Senior PC for ACML, as PC of top conferences on web search and data mining, e.g., SIGIR, WWW, NIPS, IJCAI, CIKM, WSDM, as reviewer of several leading journals and publishers, e.g., TKDE, TOIS, TIST, NOW publishers, and Springer.

Title of Speech: Reinforcement Learning to Rank for Search Result Diversification 

Abstract: The goal of search result diversification is to construct a document ranking for satisfying as many different query subtopics as possible. Typically, the diverse ranking process can be formalized as greedy sequential document selection. At each position, the document that can provide the largest amount of additional information to the users is selected. Since the utility of a document depends on its preceding documents in search result diversification, constructing an optimal document ranking is NP-hard. The traditional greedy document selection usually leads to suboptimal solutions. In the talk, I will show that the problem can be alleviated with a Monte Carlo tree search (MCTS) enhanced Markov decision process (MDP) model. Specifically, the sequential document selection process is fit into an MDP. At each time step the greedy action is further improved through the exploratory tree search by MCTS. Reinforcement learning algorithm was developed to learn the model parameters. Empirical evaluation clearly indicated the effectiveness of the approach. The MCTS enhanced MDP can also be applied to variant applications, including sequence tagging, text matching etc.  

Jane Doe
Assoc. Prof. Yong Chen, Texas Tech University, USA

Biography: Yong Chen is an Associate Professor and Director of the Data-Intensive Scalable Computing Laboratory in the Computer Science Department of Texas Tech University. He is also a Site Director of the Cloud and Autonomic Computing center at Texas Tech. His research interests include data-intensive computing, parallel and distributed computing, high-performance computing, and cloud computing. He has published over 100 research papers in international journals and conferences, and received several awards for his research activities including the IEEE TCSC (Technical Committee on Scalable Computing) Young Achievers Award, the Ralph E. Powe Junior Faculty Enhancement Award, Texas Tech University Whitacre College of Engineering Research Award, Texas Tech University Mortar Board and Omicron Delta Kappa Outstanding Faculty Award, several Best Paper Awards including The 11th IEEE International Conference on Networking, Architecture, and Storage (NAS), The 14th and the 9th IEEE International Symposium on Parallel and Distributed Processing with Applications (ISPA), The 7th International Conference on Future Information Technology, and Best Paper finalist and Best Student Paper finalist at the ACM/IEEE Supercomputing Conference (SC). His research has been funded by the National Science Foundation, Department of Defense, Department of Energy/Argonne National Laboratory, Oak Ridge Associated University, Dell Inc., Nimboxx, Jabil/Stack Velocity, and NVidia. He has also served as editors, chairs, and program committee members for numerous international journals, conferences, and workshops. More information about him can be found at http://www.myweb.ttu.edu/yonchen/.

Title of Speech: High Performance Computing Revisited for Big Data Applications  

Abstract: The increasingly important data-intensive scientific discovery presents a critical question to the high performance computing (HPC) community - how to efficiently support these growing scientific big data applications with HPC systems that are traditionally designed for big compute applications? The conventional HPC systems are computing-centric and designed for computation-intensive applications. Scientific big data applications have different characteristics compared to big compute applications. These scientific applications, however, will still largely rely on HPC systems to be solved. In this talk, we try to answer this question with a rethinking of HPC system architecture. We study and analyze a decoupled HPC system architecture for data-intensive scientific applications. The fundamental idea is to decouple conventional compute nodes and dynamically provision as data processing nodes that focus on data processing capability. We present studies and analyses for such decoupled HPC system architecture. Its data-centric model and architecture can have an impact in designing and developing future HPC systems for big data applications. 

HPCCT2016 Keynote Speakers

Jane Doe
Prof. Jun Wang, City University of Hong Kong

王钧,IEEE Fellow(国际电气和电子工程师协会会士),IEEE Fellow遴选委员会委员,IAPR Fellow(国际模式识别协会会士),香港城市大学教授,国家“千人计划”讲席教授,教育部“长江学者”讲席教授。
王钧教授的研究方向主要包括神经网络及应用,在国际一流学术刊物上发表了150多篇论文,其中半数以上论文发表在IEEE Transactions系列期刊,据不完全统计,被SCI引用4100多次。主编专著10部,参与编写专著12本,历任十几个SCI期刊的主编或副主编。研究成果曾获得2008-2009年度香港中文大学优秀成果奖,2009年度上海市自然科学成果一等奖,2010-2012年度IEEE计算智能学会杰出学者,2011年度IEEE计算智能学会优秀论文奖,2011年度亚太神经网络大会杰出成就奖,2011年度国家教育部自然科学成果一等奖等诸多奖励。


Jane Doe
Prof. Weimin Zheng, Tsinghua University, China
郑纬民教授, 清华大学

1970 年毕业于清华大学自动控制系并留校任教。1982 年获硕士学位,现为清华大学计算机系教授、博士生导师。担任中国计算机学会副理事长;中国计算机学会学术工会主任;北京市科技进步奖第八届评审委员会委员等。1985-1986 年曾在美国纽约州立大学石溪分校从事分布操作系统研究。1989-1991年曾在英国南安普敦大学参加函数语言并行编译系统研究。
主要荣誉称: 获国家科技进步一等奖 1 项,部级科技进步奖 6 项。被评为清华大学先进工作者 4 次。清华大学优秀共产党员 1 次。优秀博士后合作导师 1 次。
学科建设:作为计算机系统结构的学科带头人, 郑纬民 教授十分重视学科建设,密切关注计算机学科的发展,并能准确地把握学科的发展方向,将集群计算机、 CPU 设计、网格计算、高性能存储等确定为高性能计算技术研究所的研究方向,研究水平在国内处于领先地位。十五期间,高性能计算技术研究所共承担多项重要课题: 973 课题 2 项,国家基金委重点项目 2 项、国家基金委面上项目 12 项目、国防预研项目 3 项, 863 项目 11 项及多项国际合作项目等。主持研制的清华集群计算机系统性能多次进入同期全球高性能计算机系统 TOP 500 之列。目前清华集群计算机系统已经应用于北京、天津等省市的日常气象预报,并在网络信息安全等部门发挥了重要作用。获国家科技进步奖一等奖一项。研制的 OpenMP 编译器、检查点设置和回卷系统以及并行调试器已成为国内外多所大学进行进一步研究工作的基础。在他的带领下,用一年的时间研制成功了我国当时运行频率最高的高性能嵌入式 CPU-THUMP107 ,其最高频率为 500MHz ,功耗 <0.5W 。他领导的网格计算研究组取得了多项具有自主知识产权的成果,已推广到 20 多家重点大学使用。由他牵头完成的生物信息学网格已经投入实际运行 , 每天的用户访问量超过了 5 万人次。所研制的网格中间件、网格监控系统和网格互连互通技术达到了国际先进水平,己应用于国家科技大平台、上海科技大平台和某国防单位。目前他负责承担国家科技大平台生物信息学应用系统。
在网络存储方面 , 研制了的“高可扩展海量存储网络系统 TH-MSNS ”,已通过中国软件评测中心评测和教育部的鉴定,“该系统在整体技术上达到了国际先进水平,其中分布式虚拟存储技术和多路远程镜像技术具有重要创新”,“带外虚拟存储技术、大容量内存虚拟磁盘技术进入国际领先行列”。 TH-MSNS 系统是国内具有自主知识产权的国产网络存储系统,分别获 2004 年北京市科技二等奖、 2005 年首届“中国计算机学会创新奖”二等奖等, 2005 年国家 863 计划首届“浪潮高性能计算创新集体奖”二等奖。该系统已申请 13 项专利,已授权 4 项,获得 15 项软件著作权登记,相关研究成果发表在 IEEE Trans. on Computers 、 ACM Trans. on Storage 和 Mass Storage Systems and Technologies 等重要国际期刊和会议上。目前 TH-MSNS 系统已有 70 余套应用在国家审计署、北京市公安局、胜利油田、视频、高校等行业或部门,特别是为“金审工程”二期建设的实施提供了重要的支撑平台 , 为国家数据安全的提供了有力的保障。 郑纬民 教授连续 4 年科研经费在计算机系列第一。