Prof. Dr. Sergei Gorlatch, University of Muenster, Germany
Prof. Sergei Gorlatch has been Full Professor of Computer Science at the University of Muenster (Germany) since 2003.
Earlier he was Associate Professor at the Technical University of Berlin, Assistant Professor at the University of Passau, and
Humboldt Research Fellow at the Technical University of Munich, all in Germany. Prof. Gorlatch has about 200 peer-reviewed publications in renowned international books,
journals and conferences. He has been principal investigator in several international research and development projects in the field of parallel, distributed,
Grid and Cloud algorithms, networking and computing, as well as e-Learning, funded by the European Commission and by German national bodies.
Speech: Distributed Applications Based on Mobile Cloud Computing and Software-Defined Networks
Abstract: We consider an emerging class of challenging networked multimedia applications called Real-Time Online Interactive Applications (ROIA). ROIA are networked applications connecting a potentially very high number of users who interact with the application and with each other in real time, i.e., a response to a user’s action happens virtually immediately. Typical representatives of ROIA are multiplayer online computer games, advanced simulation-based e-learning and serious gaming. All these applications are characterized by high performance and QoS requirements, such as: short response times to user inputs (about 0.1-1.5 s); frequent state updates (up to 100 Hz); large and frequently changing numbers of users in a single application instance (up to tens of thousands simultaneous users).
This talk will address two challenging aspects of future Internet-based ROIA applications: a) using Mobile Cloud Computing for allowing high application performance when a ROIA application is accessed from multiple mobile devices, and b) managing dynamic QoS requirements of ROIA applications by employing the emerging technology of Software-Defined Networking (SDN).
Prof. James T. Kwok, IEEE Fellow, Hong Kong University of Science and Technology, Hong Kong
Prof. Kwok is a Professor in the Department of
Computer Science and Engineering, Hong Kong University of Science and
Technology. He received his B.Sc. degree in Electrical and Electronic
Engineering from the University of Hong Kong and his Ph.D. degree in computer
science from the Hong Kong University of Science and Technology. Prof. Kwok
served/is serving as an Associate Editor for the IEEE Transactions on Neural
Networks and Learning Systems, Neurocomputing and the International Journal of
Data Science and Analytics. He has also served as Program Co-chair of a number
of international conferences, and as Area Chairs in conferences such as NIPS,
ICML, ECML, AAAI and IJCAI. He is an IEEE Fellow.
Speech: Small is beautiful: The quest for compact machine learning models
Abstract: Many machine learning models, such as deep networks, are powerful but computationally expensive. This often hinders deployment to small computing devices such as cell phones and the internet of things. In this talk, we consider two approaches to compress deep learning models so that the network size can be reduced without sacrificing performance. One is to quantize each full-precision weight to a small number of bits. This can be formulated as an optimization problem which is then solved efficiently by the proximal Newton algorithm. Besides greatly reducing the network size, this can also replace the underlying multiplications to simpler bit operations. The second approach is pruning, which resembles weakening of functional connections in biological brains. We show that the pruned sparse deep networks exhibit properties of the power law, which has been observed in the degree distributions of many biological networks. The power law is also naturally related to preferential attachment, and we propose a model to explain how the deep network topology evolves in continual learning.
Prof. Nguyen Huu Thanh, Hanoi University of Science and Technology, Vietnam
Nguyen Huu Thanh received B.S and M.Sc in Electrical
Engineering from Hanoi University of Science and Technology, Vietnam in 1993 and
1995, respectively. In 2002, he received his PhD with summa cum laude in
Computer Science from the University of Federal Armed Forces Munich (Germany).
From 2002 to 2004 he has been with the Fraunhofer Institute for Open
Communication Systems (FOKUS) in Berlin, Germany. From 2004 until now Nguyen Huu
Thanh has been working as associate professor in the School of Electronics and
Telecommunications, Hanoi University of Science and Technology, Vietnam. His
research interests include radio resource management in 4G/5G systems, QoS/QoE,
network security, the Future Internet, energy-efficient networking and
software-defined networking. He is the leader and member of several national and
international research projects on network security and the Future Internet. He
is currently the Chair of IEEE Vietnam Section, executive committee member of
the Radio Electronics Association of Vietnam (REV).
Speech: Green Cloud Computing – Current Trends and Issues
Abstract: Cloud computing has emerged in the recent years as a promising paradigm that facilitates such new service models as Infrastructure-as-a-Service (IaaS), Platform-as-a-Service (PaaS) and Software-as-a-Service (SaaS). These services are becoming more common and require additional network and computing resources. In such paradigm, data centers play a crucial role as they provide the infrastructure for cloud services. Furthermore, data centers nowadays are specifically designed to be extremely over dimensioned in terms of number of servers, switching capacity and of number of deployed links and nodes to guarantee zero-loss and minimum latency packet forwarding. Recent surveys have shown that the energy consumption in a data center considerably contributes to its operation costs. A remarkable part of the large energy volume consumed in data centers today is due to the aforementioned over-provisioning to meet the stringent requirements on reliability. Therefore performance and energy issues are important factors for the design of large multi-tier data centers that can support multiple services.
The keynote is divided into two parts. In the first part of the talk, the motivations for current research activities on the areas of Green ICT are addressed. Recent hot research topics for energy-efficient networking, especially green data centers and clouds will also be discussed. In the second part, experiences and research results on the areas of Green Networking at the Future Internet research group (HUST) will be shared, including data center traffic measurement and modeling, energy profiling of network devices, energy-aware data center virtualization, energy-aware testbed, protocols and optimization algorithms.
Prof. Jianjun Li, Hangzhou Dianzi University, China
Jianjun Li received the B.S. degree in electrical
engineering from Xi’an University of Electronic Science and Technology, China in
1990 and the M.S. degree from University of Western Ontario, Canada and Ph.D
degree from University of Windsor, Canada separately. He is now working in
Hangzhou Dianzi University as a chair professor. His research interests include
artificial intelligence, computer vision, video coding, transcoding, image
processing algorithms and implementation.
Speech: Separable convolution for stereo visual matching network
Abstract: Recent work has exploited context information for finding correspondence in ill-posed regions. However current architectures rely on expensive computation to deal with ill-posed regions. To tackle this problem, we explore separable convolution and Atrous Spatial Pyramid Pooling in the application of the stereo match. The Atrous Spatial Pyramid Pooling model probes convolution features at multiple scales to take advantage of the capacity of global context information. 3D convolution was explored to learn context information while it spent expensive computation and occupy lots of GPU memory. Our proposed SCSMNet extends PSMNet by employing depthwise separable convolution with efficiently calculating and the modified ASPP model.
Prof. Phuoc Vinh Tran, Thu Dau Mot University, Vietnam
Phuoc Vinh Tran is a geographic information scientist who has been nationally recognized as Associate Professor
by the Vietnamese National Council for Professor Title since 2002. He received his PhD in Informatics from Vietnam National University in 1994.
In 1999, he was a United States National Science Foundation (NSF) fellow researcher at the University of Kentucky and the University of California – Santa Barbara.
In 2006, he was also an exchange researcher at the RMIT University in Melbourne, Australia.
Dr. Tran joined the faculty of Ho Chi Minh City University of Technology (HCMUT) in 1976. In 1994, he founded and became the director of the Center for Developing IT and GIS (DITAGIS), established and later served as the General Chair of a series of Annual International Conferences of GIS, the largest GIS events in Vietnam in 1990s and 2000s. In 1999, he was also appointed as Vice Dean of the Faculty of Environment and Natural Resources. In 2006, Dr. Tran left HCMUT to serve as Vice Rector of the University of Information Technology (UIT). In 2014, Dr. Tran retired from UIT to serve as Vice Provost of Tan Tao University (TTU); then as a specialist of Thu Dau Mot University (TDMU) and as a member of the Committee of Kien Giang University (KGU) from 2015.
Dr. Tran has more than 20 years of professional and academic experience in the field of geographic information science. He is interested in spatio-temporal data visualization, real-time spatial information system, real-time geo-visualization, real-time GIS for natural disaster and climate change, GIS for administration, and smart traffic control. He has more than 60 publications in international journals and conferences, four Vietnamese GIS books; He was also an editor of the GIS Development magazine. Dr. Tran has advised over 80 master students and 5 Ph.D students. He has also been the principal investigator of many national research and development projects; He was also a specialist of International Consultant Company CDM in a project funded by World Bank. Dr. Tran was awarded medal by MOST and medal recipient of the Vietnam Fund for Supporting Technological Creation (Vifotec).
Speech: Smart System Controlling Traffic Lights based on Direction Priority
Abstract: One of causes of repeated traffic congestion at intersections in Hochiminh City is the unreasonableness of the current system controlling traffic lights that switches lights at fixedly programmed time points. In several cases, vehicles are compelled to stop at stop line while nothing travels on front intersection, and emergency vehicles can not overpass crowd of vehicles stopping before traffic lights. Moreover, the too slow movement of vehicles within intersection due to conflict of directions is also a cause of congestion and jam. This study proposes an approach, termed the approach to priority-based controlling traffic lights, to serving as a module of smart system controlling traffic lights. The approach eliminates the conflict of vehicles moving within intersection to decrease the time passing crossroad. Traffic lights are switched according to the priority of directions to reduce the thickness of vehicles before traffic lights of directions. As a result, the approach prevents congestion within intersection and behind traffic lights at flows entering crossroad.