Liang-Gee ChenNational Taiwan UniversityMay 24, 2021 09:00-10:00 (KST) Circuits and Systems Drive Semiconductor MovingLiang-Gee Chen (陳良基) is a Professor of Electrical Engineering at National Taiwan University in Taipei.
He obtained his B.Sc., M.Sc. and Ph.D. degrees from National Cheng Kung University, Tainan, Taiwan, R.O.C. in 1979, 1981, and 1986, respectively. In 1988, he joined the Department of Electrical Engineering, National Taiwan University. During 1993–1994, he was a Visiting Consultant in the DSP Research Department, AT&T Bell Labs, Murray Hill, NJ. In 1997, he was a Visiting Scholar of the Department of Electrical Engineering, University of Washington, Seattle. During 2004-2006, he was the Vice President and General Director of the Electronics Research and Service Organization (ERSO) of the Industrial Technology Research Institute (ITRI). Since 2007, he has been serving as a Co-Director General of National SoC Program. He was the Deputy Dean of office of Research and Development in National Taiwan University during 2008-2009. During 2009-2012, he was the Deputy Dean of college of EECS and a Distinguished Professor of Department of Electrical Engineering at National Taiwan University. He was the President of National Applied Research Laboratories during 2012-2013. He was the Executive Vice President for Academics & Research of National Taiwan University during 2013-2016. He was the Political Deputy Minister of Ministry of Education, Taiwan, R.O.C. during 2016-2017. From 2017 to 2020, he was the Minister of Ministry of Science and Technology, Taiwan, R.O.C.
He has been an IEEE Fellow from 2001 for his contributions to algorithm and architecture design on video coding systems. In 2009, he was awarded TWAS Prizes and National Professorship. His research interests are DSP IC design, video signal processing and bio-signal processing. He has authored over 550 publications, 48 patents and 31 US patents.
Dr. Chen has served in editorial capacity for various journals within the circuit design and system processing community. He is a member of Phi Tau Phi and has been an elected Fellow of the U.S. National Academy of Inventors since 2016.Due to the technology innovation, the world is driving into the intelligent society. The core of the intelligent capability comes from many smart technologies. The most valuable enabling technology is the huge progressing in infrastructure technology improving HPC server, High-bandwidth communication, and low energy edge sensing and computing. All kind of social data, generated by machine as well as human being, transferred between users and service providing environments with the help of deep learning. Those data inference and make lots of incredible functions come true, for example, self-driving car, unmanned shop, automotive factory, even work-from-home.
Semiconductor is no doubt the carrier for all the data. The shrinking of semiconductor technology makes chip small enough and become invisible embedded into the world. However, it is the circuits and systems technologies provide the high-performance data processing, deliver high bandwidth transmission, and offer the low energy device for sensing and computing. Taiwan, the world-wide most important semiconductor manufacturing industry cluster, also sensed this big technology trend and demand. We initiated a very important project, called “semiconductor moon shot program”. In this presentation, the detail of the program and latest achievements will be presented. The semiconductor moonshot program was initiated by the Ministry of Science and Technology, while I served as Minister. The purpose of the program is to enhance the competitiveness of the semiconductor industry in the core technologies of artificial intelligence (AI)-powered edge computing. From 2018 to 2021, we focus on the research and development of new semiconductor processes and advanced chip systems centered on intelligent edge computing, with an eye toward gaining an advantage in the market for applications of AIoT, or AI over the internet of things. The program has six segmented technology parts: next-generation memory designs; process technologies and materials for key components of sensing devices; unmanned vehicles and augmented and virtual reality applications; and Internet of Things systems and security. So far, the program already got many exciting progressing. We believe with the success of this semiconductor moonshot program, circuits and systems technology will create a friendly AI-eco environment and keep semiconductor moving to provide an intelligent society.
Min-Goo KimExecutive Vice President, System LSI Business, Samsung ElectronicsMay 24, 2021 15:00-16:00 (KST) How Will the Evolution of Personalized Portable Systems Affect the Smart Society?In 1993, Min-Goo Kim joined Samsung Electronics, Suwon, Korea where he played a key role in developing of wireless digital communication technologies such as digital-AMPS, CDMA, and OFDMA. He received his Ph.D. degree in electronics engineering from Seoul National University (SNU), Seoul, Korea. His areas of expertise are in mobile station technologies and its implementation in efficient way. Also his research interest includes standardization of wireless communications. From 1998, he participated 3GPP standard meeting and contributed advanced technologies for UMTS standards. From 2000, he participated 3GPP2 standard meeting and contributed advanced technologies for HARQ and AMC (adaptive modulation and coding). He proposed the HARQ structure with turbo codes in 3GPP, 3GPP2, and IEEE802.16m such as “circular buffer based HARQ structure for incremental redundancy scheme with turbo codes”.
He was the recipient of the person of merit for the 20th anniversary of 2nd foundation of Samsung Electronics, 2004 Korean President Award for Excellence in Invention on the 39th National Industry Invention Day, 2006 Hae-Dong Award for Excellence in Research, 2019 Korean President Award (Industrial Medal) for Excellence in System on Chip (SOC) on the 12th Semiconductor Day, and several awards based on research and industrial merit. He was a member of NGMN (Next Generation of Mobile Networks) and LSTI (LTE/SAE Trial Initiative).
He is currently head of the SOC Development in Samsung Electronics S.LSI Division. His current research interests include 5G New Radio, power-efficient SoC technologies and next generation GPU.In COVID-19 and Un-tact era, personal and portable devices rule out the world through the social media. Computing, communications, photos & videos, financing, security, and game are merged into the smartphone. Personalized portable systems continue to evolve and are equipped with very high-performance SoC (System on Chip) composed of new and complicated processors such as CPU, GPU, NPU, DSP, and 5G modem. System’s evolution keeps increasing exponentially, e.g. LTE to 5G/6G, wider bandwidth and very low latency could come up with the change of portable devices and user's experience.
However, recently CMOS technology approach the physical limits, scaling will no longer be the sole contributor to performance improvement. In personalized portable devices, power consumption becomes a critical factor for successful implementation, and power optimization (performance per watt) is a big issue.
In the lecture, we will discuss what breakthrough in circuit and systems (CAS), and what is the most important metric in CAS. Perhaps power consumption is one of the important factors. Personalized portable system’s requirements and its technology could trigger the breakthrough of CAS.
Giacomo IndiveriUniversity of Zurich and ETH ZurichMay 25, 2021 09:00-10:00 (KST) Neuromorphic Intelligence: Brain-Inspired Processing Technologies for Extreme-Edge Use CasesGiacomo Indiveri is a dual Professor at the Faculty of Science of the University of Zurich and at Department of Information Technology and Electrical Engineering of ETH Zurich, Switzerland.
He is the director of the Institute of Neuroinformatics (INI) of the University of Zurich and ETH Zurich. He obtained an M.Sc. degree in electrical engineering and a Ph.D. degree in computer science from the University of Genoa, Italy. He was a post-doctoral research fellow in the Division of Biology at Caltech and at the Institute of Neuroinformatics of the University of Zurich and ETH Zurich. He was awarded an ERC Starting Grant on "Neuromorphic processors" in 2011 and an ERC Consolidator Grant on neuromorphic cognitive agents in 2016.
His research interests lie in the study of real and electronic neural processing systems, with a particular focus on spike-based learning and spike-based recurrent neural network dynamics. His research and development activities focus on the full custom hardware implementation of real-time sensory-motor systems using analog/digital neuromorphic circuits and emerging memory technologies.Artificial Intelligence (AI) and deep learning algorithms are revolutionizing our computing landscape, and have demonstrated impressive results in a wide range of applications. However, they still have serious shortcomings for use cases that require closed-loop interactions with the real-world. Current AI systems are still not able to compete with biological ones in tasks that involve real-time processing of sensory data and decision making in complex and noisy settings. Neuromorphic Intelligence (NI) aims to fill this gap by developing ultra-low power electronic circuits and radically different brain-inspired in-memory computing architectures. NI hardware systems implement the principles of computation observed in the nervous system by exploiting the physics of their electronic devices to directly emulate the biophysics of real neurons and synapses.
In this lecture I will present examples of NI circuits, and demonstrate applications of NI processing systems to extreme-edge use cases, that require low power, local processing of the sensed data, and that cannot afford to connect to the cloud for running AI algorithms.
Kyoung ParkVP, Memory System Research, SK hynixMay 25, 2021 15:00-16:00 (KST) Memory & Storage Challenges for Data-centric CloudKyoung Park received his MS degree in computer engineering from Chonbuk National University, Jeonju, Rep. of Korea and PhD from Korea University, Seoul, Rep. of Korea, in 1993 and 2008, respectively. He joined ETRI, Daejeon, Seoul, Rep. of Korea, in 1993, and he had worked with ETRI for 24 years. He was involved with several high performance computing architecture research project including symmetric multiprocessing system, massive parallel computing system and high‐performance cluster computing systems. He also worked in network I/O offloading architecture, big data analytics platform and large scale deep learning for vision intelligence as project leader and department manager. Since 2017, he is working with SK Hynix in Icheon, Rep. of Korea, as a leader of memory system research. Currently, his research includes next generation data center architecture, memory & storage software and new concepts of future memory solutions.
Vivienne SzeMITMay 26, 2021 09:00-10:00 (KST) Efficient Computing for AI and Robotics: From Hardware Accelerators to Algorithm DesignVivienne Sze (http://sze.mit.edu/) is an associate professor in MIT’s Department of Electrical Engineering and Computer Science and leads the Research Lab of Electronics’ Energy-Efficient Multimedia Systems research group. Her group works on computing systems that enable energy-efficient machine learning, computer vision, and video compression/processing for a wide range of applications, including autonomous navigation, digital health, and the internet of things. She is widely recognized for her leading work in these areas and has received many awards, including faculty awards from Google, Facebook, and Qualcomm, the Symposium on VLSI Circuits Best Student Paper Award, the IEEE Custom Integrated Circuits Conference Outstanding Invited Paper Award, and the IEEE Micro Top Picks Award. As a member of the Joint Collaborative Team on Video Coding, she received the Primetime Engineering Emmy Award for the development of the High-Efficiency Video Coding video compression standard. She is a co-author of the book entitled “Efficient Processing of Deep Neural Networks”.The compute demands of AI and robotics continue to rise due to the rapidly growing volume of data to be processed; the increasingly complex algorithms for higher quality of results; and the demands for energy efficiency and real-time performance. In this talk, we will discuss the design of efficient hardware accelerators and the co-design of algorithms and hardware that reduce the energy consumption while delivering real-time and robust performance for applications including deep neural networks and autonomous navigation. We will also highlight important design principles, methodology, and tools that can facilitate an effective design process.