报告题目一：Introduction to GPU Computing
报告人：John R. Cheng
报告时间：11月3号（周五）9:15-10:00
报告地点：行政楼128报告厅
报告摘要：
The High-Performance Computing (HPC) on GPU-CPU heterogeneous architectures has been becoming mainstream for achieving high performance and power efficiency, pervasive and manifest across industry and academia throughout the world. GPU computing has led to a fundamental paradigm shift in the parallel programming world, which, together with machine learning methods, is regarded as the two core technologies of the AI computing age. This talk will introduce you the wonderful world of heterogeneous parallel programming with CUDA C!
This talk will briefly cover the following topics:
1. The paradigm of Heterogeneous Parallel Programming
2. CUDA 9: the powerful platform for GPU-accelerated applications.
3. Professional CUDA C Programming: the basics
报告人简介：
Dr. John Runwei Cheng is a senior research scientist in BGP, Houston, as a worldwide well-known expert in GPU programming. PhD in Engineering, Tokyo Institute of Technology in 1996. John has been worked in the finance industry many years before he joined in oil and gas industry, as an expert in computational intelligence, to provide advanced solutions based on genetic algorithms hybridized with data mining and statistical learning to solve real world business challenges. As an internationally recognized researcher in the field of genetic algorithms and their application to industrial engineering, John has co-authored three books with Prof. Mitsuo Gen. John has a wide range of experience in both academic research and industry development, and is gifted in making complex subjects accessible to readers with a concise, illustrative, and edifying approach.
报告题目二：Accelerating AI with GPUs: The State-of-the-Art
报告人：Michael Ma
报告时间：11月3号（周五）10:00-10:45
报告地点：行政楼128报告厅
报告摘要：
AI is everywhere. It is more ubiquitous than most realize.
AI is touching our lives and bridging barriers in our communities
AI has no boundaries. Every company has data and every company needs intelligence.
Today, we stand at the beginning of the next era, the AI computing era, ignited by a new computing model, GPU deep learning. This new model — where deep neural networks are trained to recognize patterns from massive amounts of data — has proven to be “unreasonably” effective at solving some of the most complex problems in computer science. In this era, software writes itself and machines learn. Soon, hundreds of billions of devices will be infused with intelligence. AI will revolutionize every industry.
Briefly cover the following topics:
Tesla GPU products overview
Tesla acceleration AI deep learning
AI supercomputer DGX
报告人简介：
Michael Ma is currently responsible for the Education and Research Industry Business Development in NVIDIA. He previously worked for Oracle for five years.
报告题目三：Hybrid Evolutionary Optimization with Learning for Scheduling
报告人：Lin Lin
报告时间：11月3号（周五）11:00-11:45
报告地点：行政楼128报告厅
报告摘要：
Evolutionary Algorithms (EAs) has attracted significantly attention with respect to complexity scheduling problems, which is referred to evolutionary scheduling. However, EAs differ in the implementation details and the nature of the particular scheduling problem applied. In order to have an effective implementation of EAs for production scheduling, this paper focuses on making a survey of researches based on using hybrid EAs. Starting from scheduling description, we identify the classification and graph representation of scheduling problems. Then we present the various representations, hybridization techniques, and machine-learning techniques to enhancing EAs. Finally, we also present successful applications in manufacturing.
报告人简介：
Dr. Lin Lin is currently an Associate Professor at School of Software, Dalian University of Technology (DLUT), China; and is a senior researcher at Fuzzy Logic Systems Institute, Japan. He received his M.Sc degree and Ph.D. degree in Engineering from Graduate School of Information, Production and Systems, Waseda University in March 2005 and March 2008, respectively. He was a Research Assistant at Information, Production and Systems Research Center (IPSRC), Waseda University from April 2006 to March 2008, and was a Visiting Lecturer at IPSRC, Waseda University, as a Postdoctoral Research Associate supported by the Kitakyushu Foundation for the Advancement of Industry, Science and Technology (FAIS). He is a coauthor of Network Models and Optimization: Multiobjective Genetic Algorithm Approach, Springer-Verlag in 2008. His core research interests are computational intelligence, deep learning, probabilistic graphical models, and their applications in combinatorial optimization and pattern recognition.
报告题目四：Study on the reliability evaluation model and method of underground container logistics system under the combined development of port city
报告人：Yi-qun Fan
报告时间：11月3号（周五）11:45-12:30
报告地点：行政楼128报告厅
报告摘要：
This talk focuses on the negative effect of container transportation from Waigaoqiao Port in Shanghai on the traffic of surrounding area. By conducting comprehensive cost-benefit analysis of several aspects including engineering investment, risk and environmental issues, Urban Container Freight Transportation (UCFT) is recommended in this case, and conceptual design of UCFT is illustrated. Then the simulation model of container terminal of container underground logistics system is studied and the simulation analysis is carried out. The research shows that the system not only includes visual simulation, and includes system simulation, namely on the operational parameters, route of each unit in the system and so on are all in the detailed design to achieve the vivid result ,which have strong authenticity, rationality at the same time.
报告人简介：
Dr. Fan Yi-qun is the deputy chief engineer and the director of R&D Center of Urban Traffic and Underground Space Design Institute, SMED
