Distinguished Lectures Series: 2017-2018 Speakers

Dr. Stephen W. Keckler, NVIDIA

Thursday, Sept. 28, 2017
4:00 p.m., room SF1105

TitleConsumer-focused High Performance Computing Architectures

Abstract: The technology landscape is incredibly exciting today, with high-performance computation transforming many aspects of society and daily life. New innovations appear seemingly daily in areas of entertainment, transportation, communication, and health care, just to name a few.  Emerging practical applications of virtual and augmented reality, autonomous vehicles, and automated reasoning will place new demands on our computing architectures.  While the computational appetite of emerging applications in these spaces appear to be growing without bound, the historical technology scaling trends which have provided the fundamental horsepower for computing over the last 50 years, are slowing substantially.  This talk will discuss some of the cataclysmic trends in consumer applications of high-performance computing, and focus on opportunities for computer designers. It will also present challenges associated with emerging deep neural networks (DNNs) and describe recent works that (1) enable larger and more complex networks to be trained on compute devices with limited memory capacity; and (2) reduce the memory and computation footprints of DNNs at inference time, enabling them to run with vastly improved energy efficiency.

BioDr. Stephen W. Keckler is the Vice President of Architecture Research at NVIDIA and an Adjunct Professor of Computer Science at the University of Texas at Austin, where he served on the faculty from 1998-2012. His research interests include parallel computer architectures, high-performance computing, energy-efficient architectures, and embedded computing.  Dr. Keckler is a Fellow of the ACM, a Fellow of the IEEE, an Alfred P. Sloan Research Fellow, and a recipient of the NSF CAREER award, the ACM Grace Murray Hopper award, the President’s Associates Teaching Excellence Award at UT-Austin, and the Edith and Peter O’Donnell award for Engineering. He earned a B.S. in Electrical Engineering from Stanford University and M.S. and Ph.D. degrees in Computer Science from the Massachusetts Institute of Technology.

 

Professor Daniël De Zutter, University of Ghent

Thursday, Oct. 5, 2017
4:00 p.m., room SF1105

Title: Broadband Electromagnetic Modelling and Stochastic Signal Analysis of Multiconductor Interconnections

 AbstractThis lecture addresses the statistical modelling and simulation of high-speed interconnections with uncertain physical properties and terminations. Typical quantities of interest are S-parameters, crosstalk, Bit Error Rates and eye-diagrams. Focus is on on-board and on-chip multiconductor interconnections modelled by the telegrapher equations. First, the meaning of the classical resistance, inductance, conductance and capacitance (RLGC) per unit-of-length parameters in the presence of good conductors and semiconductors, is revisited. Next, attention is devoted to an efficient numerical approach to obtain broadband RLGC-data for arbitrary cross-sections, while accurately taking skin-effect and current crowding into account. This efficient numerical technique lays the foundation for the subsequent statistical signal analysis where stochastic variations in the physical link properties and in the load conditions are included. It is shown how Polynomial Chaos based approaches can be advantageously harnessed to beat the traditional Monte Carlo technique and how these techniques can be integrated in SPICE-compatible simulators. Finally, we consider the case in which only a limited set of measurements or simulations of a passive interconnection, exhibiting variability, are available, without any further physical or geometrical insight. It is investigated how machine learning techniques allow to create a realistic and physically consistent stochastic model of that interconnection and its signal transfer properties.

BioDaniël De Zutter was born in 1953. He received his M. Sc. Degree in Electrical Engineering from the University of Gent in 1976. In 1981 he obtained a Ph. D. degree and in 1984 he completed a thesis leading to a degree equivalent to the French Aggrégation or the German Habilitation. He is a full professor of electromagnetics. His research focuses on all aspects of circuit and electromagnetic modelling of high-speed and high-frequency interconnections and packaging, on Electromagnetic Compatibility (EMC) and numerical solutions of Maxwell’s equations. As author or co-author he has contributed to more than 250 international journal papers and 270 papers in conference proceedings. In 2000 he was elected to the grade of Fellow of the IEEE. Between 2004 and 2008 he served as the Dean of the Faculty of Engineering and Architecture of Ghent University and was the head of the Department of Information Technology and of its Electromagnetic Research group until February 2017.

 

Professor Jelena Kovačević, Carnegie Mellon University

Thursday, Nov. 2, 2017
4:00 p.m., room SF1105

Title: From Biomedical Imaging to Online Blogs: Graph Signal Processing

 AbstractI will present a path from classification in biomedical imaging to online blogs, where a common thread is graph signal processing, a theoretical framework that generalizes fundamental concepts of classical signal processing from regular domains, such as lines and rectangular lattices, to general graphs. It is particularly applicable to domains such as physical, engineering, and social, where signals are characterized by irregular structure. Signal processing on graphs has found multiple applications, including approximation, sampling, classification, inpainting and clustering, and I will describe some of these.

Bio: Jelena Kovačević received a Ph.D. degree from Columbia University. She then joined Bell Labs, followed by Carnegie Mellon University in 2003, where she is currently the Hamerschlag University Professor and Head of the Department of ECE, and Professor of BME. She received the Dowd Fellowship at CMU, Belgrade October Prize, and the E.I. Jury Award at Columbia University. She is a coauthor on an SP Society award-winning paper and is a coauthor of the textbooks Wavelets and Subband Coding and Foundations of Signal Processing. Dr. Kovacevic is the Fellow of the IEEE and was the Editor-in-Chief of the IEEE Transactions on Image Processing. She was a keynote speaker at a number of meetings and has been involved in organizing numerous conferences. Her research interests include multiresolution techniques, graphs, biomedical imaging, and smart infrastructure.

 

Professor Francesco Bullo, University of California, Santa Barbara

Thursday, Nov. 30, 2017
4:00 p.m., room SF1105

Title: On the Dynamics of Influence and Appraisal Networks

 AbstractThis talk will present models for the evolution of interpersonal influences, interpersonal appraisals, and social power in a group of individuals. Specifically, we will propose learning models in two scenarios: groups who discuss and form opinions along a sequence of issues, and groups who execute a sequence of decomposable tasks.  In both scenarios we establish the emergence of rational optimal behavior, or lack thereof, as a result of the natural dynamical evolution of interpersonal appraisals and influence structures.  Our multiagent models and analysis results are grounded in influence networks from mathematical sociology, replicator dynamics from evolutionary games, and transactive memory systems from organization science.

BioFrancesco Bullo is a Professor with the Mechanical Engineering Department and the Center for Control, Dynamical Systems and Computation at the University of California, Santa Barbara. His research interests focus on network systems and distributed control with application to robotic coordination, power grids and social networks. He is the coauthor of “Geometric Control of Mechanical Systems” (Springer, 2004) and “Distributed Control of Robotic Networks” (Princeton, 2009); his forthcoming “Lectures on Network Systems” is available on his website. He received best paper awards for his work in IEEE Control Systems, Automatica, SIAM Journal on Control and Optimization, IEEE Transactions on Circuits and Systems, and IEEE Transactions on Control of Network Systems. He is a Fellow of IEEE and IFAC. He has served on the editorial boards of IEEE, SIAM, and ESAIM journals, and will serve as IEEE CSS President in 2018.

 

Dr. Bogdan Kasztenny, Schweitzer Engineering Laboratories Inc.

Thursday, Jan. 18, 2018
4:00 p.m., room SF1105

Title: Breaking the Speed Barrier of Today’s Line Protective Relays

 AbstractToday’s electric power grids interconnect an increasing number of inverter-based sources fed from renewable energy resources: wind turbines and solar panels. DC links are being planned to transport vast amounts of energy across continents linking multiple AC grids. As a result of these changes, today’s grids are more difficult to protect against short-circuit currents and other abnormal events than in the past.

Traditional protection devices, such as transmission line relays, respond to narrow-band-filtered voltages and currents, and thus—for detecting short-circuits—they count on the power sources to supply fault current toward the short-circuit location. In its 100-year history, the engineering field of power system protection has developed sophisticated protection principles. These principles, however, take advantage of the many properties of a traditional source—a large synchronous generator. When applied near inverter-based sources, these principles are tested and may underperform or fail. The industry is going through an adjustment phase, working on protection solutions that do not depend on the fixed properties of a synchronous generator, but work with inverter-based sources that use arbitrary and proprietary control schemes when generating output currents.

This lecture focuses on novel protection principles based on traveling waves and incremental signals. These protection principles respond to transients fed with the energy stored in the grid prior to a short-circuit, and therefore they are much less dependent on the sources’ outputs during the short circuit.

These protection principles offer dramatic improvements in the operating times and sensitivity. They are good examples of a disruptive technology. This lecture introduces several of the new principles and explains why only now these principles can be fully developed and can make their way into practice. These enablers include high-speed analog-to-digital converters and vast processing power afforded at industrial-grade component ratings, high-speed communications, precise wide-area timing, and modern simulation tools. The lecture will use actual field cases and actual devices to demonstrate the principles.

BioDr. Bogdan Kasztenny is a Senior Engineering Director in R&D at Schweitzer Engineering Laboratories, Inc. He has over 27 years of expertise in power system protection and control, including 10 years of academic career and 17 years of industrial experience, developing, promoting, and supporting many protection and control products. Bogdan is an IEEE Fellow, Senior Fulbright Fellow, Canadian representative of CIGRE Study Committee B5, and a registered professional engineer in the province of Ontario. Bogdan serves on the Western Protective Relay Conference Program Committee and on the Developments in Power System Protection Conference Program Committee. Bogdan has authored over 200 technical papers and holds over 30 patents. Since 2014, Bogdan has been heavily involved in research, development, and support of protective relays and fault locators based on traveling waves and superimposed components.

 

Professor Manfred Morari, University of Pennsylvania

Thursday, Feb. 1, 2018
4:00 p.m., room SF1105

Title: The Past, Present and Future of Control

 AbstractReflecting on our work over the last 40 years I found that it was dominated by two themes: computation and uncertainty. I will describe how the rapidly increasing computational resources have affected our approaches to deal with uncertainty in feedback control. The talk will be illustrated by examples from process control and other application areas like automotive and power systems.

BioManfred Morari was head of the Department of Information Technology and Electrical Engineering at ETH Zurich from 2009 to 2012 and head of the Automatic Control Laboratory from 1994 to 2008. Before that he was the McCollum-Corcoran Professor of Chemical Engineering and Executive Officer for Control and Dynamical Systems at the California Institute of Technology. From 1977 to 1983 he was on the faculty of the University of Wisconsin. He obtained the diploma from ETH Zurich and the Ph.D. from the University of Minnesota, both in chemical engineering. His interests are in constrained and robust control. Morari’s research is internationally recognized. The analysis techniques and software developed in his group are used in universities and industry throughout the world. He has received numerous awards, including the Eckman Award, Ragazzini Award and Bellman Control Heritage Award from the American Automatic Control Council; the Colburn Award, Professional Progress Award and CAST Division Award from the American Institute of Chemical Engineers; the Control Systems Technical Field Award and the Bode Lecture Prize from IEEE. He is a Fellow of IEEE, AIChE and IFAC. In 1993 he was elected to the U.S. National Academy of Engineering, in 2015to the UK Royal Academy of Engineering. Manfred Morari served on the technical advisory boards of several major corporations.

 

Professor Rinaldo Castello, University of Pavia

Thursday, Feb. 8, 2018
4:00 p.m., room SF1105

Title: Wireless Transceivers: Architectural and Circuit Evolution toward a Complete Integrated Solution

 AbstractAs we move toward 5G, improvements by 1,000x in system capacity and 10x in data rate, energy and spectral efficiency are required to satisfy the projected demands. To this end, new features will be introduced (e.g. un-licensed bands re-use, channel bonding, Full Duplex) and existing ones (e.g. MIMO, Carrier Aggregation) will be enhanced. This will drastically increase terminal complexity while at the same time lower cost, size and power consumption will be required. Within this scenario, the only way to keep meeting the expectations of the market is by increasing the level of integration of the terminal. This talk analyzes the progresses already achieved and those expected with respect to the RF transceiver. First the evolution of the architecture of a multi-standard transceiver up to the present time associated with the CMOS technology scaling. After, a discussion on the architecture of future transceiver capable to satisfy the requirements of the new standard like Saw/Duplexer-less implementation, full-duplex architecture, very broadband channels etc. will be provided.

BioRinaldo Castello (S’78–M’78–SM’92–F’99) graduated from the University of Genova (summa cum laude) in 1977 and received the M.S. and the Ph. D. from the University of California, Berkeley, in ‘81 and ‘84. From ‘83 to ‘85 he was Visiting Assistant Professor at the University of California, Berkeley. In 1987 he joined the University of Pavia where he is now a Full Professor. He consulted for ST-Microelectronics, Milan, Italy up to 2005 in ‘98 he started a joint research centre between the University of Pavia and ST and was its Scientific Director up to ‘05. He promoted the establishing of several design centre from multinational IC companies around Pavia, among them Marvell for which he was a consultant from 2005 to 2016. He is now consulting for InvenSense. Rinaldo Castello has been a member of the TPC of the European Solid State Circuit Conference (ESSCIRC) since 1987 and of the International Solid State Circuit Conference (ISSCC) from ‘92 to ‘04. He was Technical Chairman of ESSCIRC ’91 and General Chairman of ESSCIRC ‘02, Associate Editor for Europe of the IEEE J. of Solid-State Circ. from ’94 to ’96 and Guest Editor of its July ’92 special issue. From 2000 to 2007 he has been Distinguished Lecturer of the IEEE Solid State Circuit Society. Prof Castello was named one of the outstanding contributors for the first 50 and 60 years of ISSCC and a co-recipient of the Best Paper Award at the 2005 Symposium on VLSI of the Best Invited Paper Award at the 2011 CICC and of the Best Evening Panel Award at ISSCC 2012 and 2015. He was one of the two European representatives at the Plenary Distinguished Panel of ISSCC 2013 and the Summer 2014 Issue of the IEEE Solid State Circuit Magazine was devoted to him. Rinaldo Castello is a Fellow of the IEEE.

 

Professor David Reitze, University of Florida

Thursday, April 5, 2018
4:00 p.m., room SF1105

Title: Black Holes Last Tango: LIGO and the Dawn of Gravitational-wave Astronomy

 Abstract:  The first direct detections of gravitational waves in late 2015 were made possible by a dedicated forty year quest to design, build, and operate LIGO, the Laser Interferometer Gravitational-wave Observatory.  In this talk, I’ll give an introduction to gravitational waves and what makes them so difficult to detect and at the same time such powerful and unique probes of the universe.   I’ll also present how we detect gravitational waves using fantastically sensitive interferometers, and talk about the first detections of colliding black holes and the potentially profound implications.  Time permitting, I’ll give a preview of where LIGO intends to go in the next decade and beyond.

BioDavid Reitze holds joint positions as the Executive Director of the LIGO Laboratory at the Caltech and a Professor of Physics at the University of Florida.  He has authored more than 250 publications, and is a Fellow of both the American Physical Society and the Optical Society. He is a member of the LIGO Scientific Collaboration (LSC) that was awarded the 2016 Special Breakthrough Prize in Fundamental Physics, the 2016 Gruber Foundation Cosmology Prize, and the 2017 Rossi Prize of the High Energy Astrophysics Division of the American Astronomical Society.  He served as the elected Spokesperson of the LSC from 2007-2011, and was recently awarded the US National Academy of Sciences Award for Scientific Discovery.