Loading Events
Find Events

Event Views Navigation

Past Events

Events List Navigation

April 2017
Free

CESG Eminent Scholar Series: “CPU and Server System Architecture Opportunities for AI Application Optimization”

April 21 @ 4:10 pm - 5:10 pm
WEB, Room 236-C,
Wisenbaker Engineering Building

CESG Eminent Scholar Series: Balint Fleischer of Huawei’s Central Research Institute “CPU and Server System Architecture Opportunities for  AI Application Optimization” Abstract:For the past 50 years, the computer industry has been focusing on improving transactional workloads. We are now seeing the emergence of a new class of “Narrow AI” based on applications playing an increasingly critical role in diverse use cases from Robotics, Smart Cities, Expert Systems, Medical Diagnostics, Financial Systems to Research and so forth. They perform assistive functions through speech recognition, face and image recognition, Fraud Detection, retrieving complex data structures and the integration of diverse information. AI applications are fundamentally different from classic applications. Classic applications are based on explicit programming using arithmetic and logic operations, while AI applications are trainable or self-learning algorithms to make predictions. AI applications use heterogeneous streaming data as opposed to classic applications, which transactional and structured data. Classic CPU architectures are very inefficient for AI applications; they lack sufficient memory BW for a diverse set of accelerators to emerge. However, E2E application “pipelines” are a hybrid requiring the creation of a new server platform capable of efficiently supporting new use cases. This presentation will highlight some of the ongoing development in this area and what could be the future direction. Bio: Balint Fleischer is currently Chief Scientist at Huawei’s Central Research Institute, where he is responsible for research into next generation data center and server architectures. He was most recently CTO at startup Parallel Machines, where he developed new architectures for advancing predictive analytics and machine learning. Previously he was the General Manager and Director of Architecture development, including efforts related to 3DXPoint and Rack Scale Architecture. He also had a long residency at Sun Microsystems including being VP/CTO of the Networked Storage Division, where he led the design of next generation storage systems and storage virtualization platforms; while at Sun he led Sun’s architecture development for many successful low end midrange server products and was responsible for the company’s InfiniBrand effort focusing on enterprise clustering, I/O, and storage. Free Snacks

Find out more »
Free

CESG Seminar: “How Much Time, Energy, and Power Does an Algorithm Need?”

April 7 @ 4:10 am - 5:10 pm
WEB, Room 236-C,
Wisenbaker Engineering Building
+ Google Map

Richard (Rich) Vuduc of Georgia Tech   Abstract: Given an algorithm and a computer system, can we estimate or bound the amount of physical energy (Joules) or power (Watts) it might require, in the same way that we do for time and storage? These physical measures of performance are relevant to nearly every class of computing device, from embedded mobile systems to power-constrained datacenters and supercomputers. Armed with models of such measures, we can try to answer many interesting questions. For instance, can algorithmic knobs be used to control energy or power as the algorithm runs? How might systems be better balanced in energy or power for certain classes of algorithms? This talk is about general ideas of what such analyses and models might look like, giving both theoretical predictions and early empirical validation of our algorithmic energy and power models on real software and systems. Bio: Rich Vuduc is an Associate Professor at the Georgia Institute of Technology (“Georgia Tech”), in the School of Computational Science and Engineering, a department devoted to the study of computer-based modeling and simulation of natural and engineered systems. His research lab, the HPC Garage (@hpcgarage), is interested in high-performance computing, with an emphasis on performance analysis and performance engineering. He has received a DARPA Computer Science Study Group grant; an NSF CAREER award; a collaborative Gordon Bell Prize in 2010; Lockheed Martin’s Award for Excellence in Teaching (2013); Best Paper Awards at the SIAM Conference on Data Mining (SDM, 2012) and the IEEE Parallel and Distributed Processing Symposium (IPDPS, 2015) among others. He also served as his department’s Associate Chair and Director of its graduate programs from 2013-2016. External to Georgia Tech, he was elected to be Vice President of the SIAM Activity Group on Supercomputing (2016-2018); co-chaired the Technical Papers Program of the “Supercomputing” (SC) Conference in 2016; and serves as an associate editor of both the International Journal of High-Performance Computing Applications (IJHPCA) and IEEE Transactions on Parallel and Distributed Systems (TPDS). He received his Ph.D. in Computer Science from the University of California, Berkeley, and was a postdoctoral scholar at the Lawrence Livermore National Laboratory.

Find out more »
Free

CESG Teleseminar: “Finite Blocklength Converses in Point-To-Point and Network Information Theory: A Convex Analytic Perspective”

April 6 @ 2:30 pm - 3:30 pm

Ankur Kulkarni – Assistant Professor Systems and Control Engineering Group Indian Institute of Technology Bombay (IITB)  Abstract: Finite blocklength converses in information theory have been discovered for several loss criteria using a variety of arguments. What is perhaps unsatisfactory is the absence of a common framework using which converses could be found for any loss criterion. We present a linear programming based framework for obtaining converses for finite blocklength lossy joint source-channel coding problems. The framework applies for any loss criterion, generalizes certain previously known converses, and also extends to multi-terminal settings. The finite blocklength problem is posed equivalently as a nonconvex optimization problem and using a lift-and-project-like method, a close but tractable LP relaxation of this problem is derived. Lower bounds on the original problem are obtained by the construction of feasible points for the dual of this LP relaxation. A particular application of this approach leads to new converses that improve on the converses of Kostina and Verdu ́ for joint source-channel coding and lossy source-coding, and imply the converse of Polyanksiy, Poor and Verdu for channel coding. Another construction leads to a new general converse for finite blocklength joint source-channel coding for a class of source-channel pairs. Employing this converse shows that the LP is tight for all blocklengths for the “matched setting” of minimization of the expected average bit-wise Hamming distortion of a q-ary uniform source over a q-ary symmetric memoryless channel. In the multi-terminal setting, using the above method we derive improvements to converses of Han for Slepian-Wolf coding, a new converse for the multiple access channel, and an improvement to a converse of Zhou et al for the successive refinement problem. Coincidentally, the recent past has seen a spurt of results on using duality to obtain outer bounds in combinatorial coding theory (including the author’s own nonasymptotic upper bounds for zero-error codes for the deletion channel). We speculate that these, and our results, hold the promise of a unified, duality-based theory of converses for problems in information theory. This is joint-work with Ph.D. student Sharu Theresa Jose. Bio: Ankur is an Assistant Professor (since 2013) with the Systems and Control Engineering group at Indian Institute of Technology Bombay (IITB). He received his B.Tech. in Aerospace Engineering from IITB in 2006 and his M.S. in 2008 and Ph.D. in 2010 – both from the University of Illinois at Urbana-Champaign (UIUC). From 2010-2012 he was a post-doctoral researcher at the Coordinated Science Laboratory at UIUC. His research interests include the role of information in stochastic control, game theory, information theory, combinatorial coding theory problems, optimization and variational inequalities, and operations research. He is an Associate (from 2015–2018) of the Indian Academy of Sciences, Bangalore, a recipient of the INSPIRE Faculty Award of the Department of Science and Technology, Government of India, 2013, the best paper award at the National Conference on Communications, 2017 and the William A. Chittenden Award, 2008 at UIUC.  He is a consultant to the Securities and Exchange Board of India on some matters…

Find out more »
March 2017
Free

CESG Seminar: “Prototyping Medium Access Control Protocols for Wireless Networks”

March 31 @ 4:10 pm - 5:10 pm
WEB, Room 236-C,
Wisenbaker Engineering Building

 Simon Yau – Texas A & M University   Abstract: Due to increasingly dense wireless deployments, increasing demand for multimedia applications, and plans to offload cellular traffic onto unlicensed bands, the efficiency of Medium Access Control (MAC) protocols has become critical to the performance of wireless networks. While many MAC protocols have been proposed, very few have been experimentally evaluated to establish realistic performance. Easy experimental evaluation of MAC protocols requires a flexible platform that is readily capable of implementing a wide range of protocols. MAC protocols have very strict timing requirements, which leads to tight coupling between the protocols and the underlying hardware. In this talk, we will present a platform for prototyping MAC protocols that uses a Mechanism vs Policy separation architecture which allows researchers to rapidly prototype different classes of MAC protocols, as well as some of the issues related to prototyping MAC protocols. Bio: Simon Yau is a Computer Engineering PhD candidate at the Department of Electrical and Computer Engineering at Texas A&M University. His research interests are in prototyping Medium Access Control protocols for current generation and next generation wireless networks. He has worked on several projects with National Instruments (NI) where some of his work is used in their 802.11 Application Framework. He currently leads the team for developing WiMAC, a rapid prototyping platform for MAC protocols and has given demonstrations of the platform at NI’s annual conference, NI Week, and SIGCOMM ’15. In addition to that, he is also currently working on developing an Unmanned Traffic Management system for

Find out more »
Free

CESG TELESEMINAR: “Collaborative Road Freight Transport”

March 30 @ 2:30 pm - 4:00 pm

“Collaborative Road Freight Transport” Karl H. Johansson –  KTH Royal Institute of Technology   Abstract: Freight transportation is of outmost importance for our society. Road transporting accounts for about 26% of all energy consumption and 18% of greenhouse gas emissions in the European Union. Goods transport in the EU amounts to 3.5 trillion ton-km per year with 3 million people employed in this sector, whereas people transport amounts to 6.5 trillion passenger-km with 2 million employees. Despite the influence the transportation system has on our energy consumption and the environment, individual long-haulage trucks with no real-time coordination or global optimization mainly do road goods transportation. In this talk, we will discuss how modern information and communication technology supports cyber-physical transportation system architecture with an integrated logistic system coordinating fleets of trucks traveling together in vehicle platoons. From the reduced air drag, platooning trucks traveling close together can save more than 10% of their fuel consumption. Control and estimation challenges and solutions on various level of this transportation system will be presented. It will be argued that a system architecture utilizing vehicle-to-vehicle and vehicle-to-infrastructure communication enable optimal and safe control of individual trucks as well as optimized vehicle fleet collaborations and new markets. Extensive experiments done on European highways will illustrate system performance and safety requirements. The presentation will be based on joint work over the last ten years with collaborators at KTH and at the truck manufacturer Scania.   Bio: Karl H. Johansson is Director of the Stockholm Strategic Research Area ICT The Next Generation and Professor at the School of Electrical Engineering, KTH Royal Institute of Technology. He received MSc and PhD degrees in Electrical Engineering from Lund University. He has held visiting positions at UC Berkeley, Caltech, NTU, HKUST Institute of Advanced Studies, and NTNU. His research interests are in networked control systems, cyber-physical systems, and applications in transportation, energy, and automation. He is a member of the IEEE Control Systems Society Board of Governors and the European Control Association Council. He has received several best paper awards and other distinctions, including a ten-year Wallenberg Scholar Grant, a Senior Researcher Position with the Swedish Research Council, and the Future Research Leader Award from the Swedish Foundation for Strategic Research. He is Fellow of the IEEE and IEEE Distinguished Lecturer.

Find out more »
Free

CESG Seminar: Exo-Core — Software-Defined Hardware-Security

March 24 @ 4:10 am - 5:10 pm
WEB, Room 236-C,
Wisenbaker Engineering Building

Mohit Tiwari – University of Texas at Austin Abstract: Confinement is a fundamental security primitive. The ability to put private data in a box and ship the box to run untrusted code in an untrusted data center can transform systems security and expand the use of cloud services to regulated data. However, untrusted applications are hard to confine — we show that using only meta-data about the computation, a malicious process can leak secrets at hundreds of kilo-bits per second on machines today. Closing such leaks in the past has followed a piece-meal approach of closing individual channels. In this talk, we propose that exposing the micro-architecture to software can enable flexible defenses to a large class of vulnerabilities, and show that software solutions can implement efficient and verifiable solutions to hardware-security problems.   Bio: Mohit Tiwari received his PhD from UCSB (2011) and joined UT Austin as an Assistant Professor in Fall 2013. His research enables privacy for end-users through information leak-free containers — such containers can be used to create trustworthy computing services using untrusted data centers and vulnerable applications — and through anomaly detection across the computing stack. Professor Tiwari’s research has received the NSF Career Award (2015), Best Paper Awards (ASPLOS’15, PACT’09), IEEE Micro Top Picks (2010, 2014 Honorable Mention), and industry research awards from Google and Qualcomm.

Find out more »
Free

CESG Teleseminar: From Brain Biometrics to Brain Hacking: Convergence of Neuroscience and Cyber Technology

March 23 @ 3:00 pm - 4:00 pm

Zhanpeng Jin Department of ECE, Binghamton University Room 333 Wisenbaker Engineering Building (fishbowl)   Abstract: Cryptographic systems often rely on the secrecy of cryptographic credentials; however, these are vulnerable to eavesdropping and can resist neither a user’s intentional disclosure nor coercion attacks where the user is forced to reveal the credentials. Conventional biometric keys (e.g., fingerprint, iris, etc.), unfortunately, can still be surreptitiously duplicated or adversely revealed. To this end, we argue that the most secure cryptographic credentials are ones of which the users aren’t even aware. On the basis of this argument, our research seeks to investigate a new psychophysiological approach for secure and trustworthy user authentication via reproducible, unique, non-volitional components of the electroencephalogram (EEG) brainwave responses, named “brainprints.” Moreover, we systematically evaluate how robust the cognitive brainprinting is to various cyber-attacks, particularly psychological and computational vulnerabilities. The preliminary results have proved the resistance of the brainprint authentication system to brainwave entrainment and impersonation. This research holds the potential to transform existing authentication systems into more secure, disclosure-resistant solutions; critical for high-security applications, as well as to strengthen our understanding of the unique cognitive and psychological secret of the human brain. “Brainprint” research has been reported by over 50 media outlets and named as one of the “future technology: 22 ideas about to change our world.”   Bio: Dr. Zhanpeng Jin is an Assistant Professor in Departments of Electrical and Computer Engineering, and Biomedical Engineering, and Director of Cyber-Med Lab at State University of New York at Binghamton. Prior to joining SUNY-Binghamton, He was a Postdoctoral Research Associate at the University of Illinois at Urbana-Champaign (UIUC) and received his Ph.D. degree in Electrical Engineering at the University of Pittsburgh. His research interests include emerging biometrics, cognitive neuroscience, cyber-physical security, neuromorphic computing, mobile health, and low-power sensing. He has served as the Associate Editor for the journals of Computers and Electrical Engineering, Computers in Biology and Medicine, and BioMedical Engineering Online, as well as served on the Technical Committees for many conferences. He received the BU ECE Outstanding Faculty Researcher Award in 2015, Best Paper Award in BHI’17, and Best Paper Award Nominee in ASP-DAC’17. His research has been supported by NSF, AFOSR, AFRL, SUNY Research Foundation, and a number of industrial companies. He is a senior member of IEEE.

Find out more »
Free

ECE Leaders and Innovators Speaker Series and TEES Tech Talk

March 10 @ 11:30 am - 12:30 pm
WEB, Room 236-C,
Wisenbaker Engineering Building

New Frontiers of Information Networks: Opportunities & Challenges Dr. John D. Matyjas Air Force Research Laboratory Information Directorate Abstract: After a brief orientation on the Air Force Research Laboratory, this talk will focus on the innovation, development, and maturation of secure communications,  networking, and information management technologies. A timely, reliable, and mission-responsive Air Force network is critical to the translation of sensory data into actionable information and for assuring tailored communications globally. To build future elastic network capabilities that can respond to the mission and threat environment, we cannot rely solely on a data-neutral network. The future lies in affordable, extensible, interoperable communications architectures that intelligently distribute information in a robust way and enable shared situational awareness and timely decision-making, ultimately, to assure the mission. These desired attributes will be discussed in the context of broadly parallel consumer and industry demands for autonomous vehicle (ground and airborne) operations and human/machine-to-machine communications. Bio: Dr. John D. Matyjas received his Ph.D. in electrical engineering from State University of New York at Buffalo in 2004. Currently, he is serving as the Tech Advisor of the Computing & Communications Division at the Air Force Research Laboratory (AFRL) in Rome, NY. His research interests include dynamic multiple-access communications and networking, software defined RF, spectrum mutability, statistical signal processing and optimization, and neural networks. Dr. Matyjas was inducted as an AFRL Fellow in 2016. He is the recipient of the 2015 Air Force Association ‘Technology Manager of the Year’ Award, 2015 AFRL ‘Scientist of the Year’ Award, 2012 IEEE R1 Technology Innovation Award, and the 2010 IEEE Int’l Communications Conf. Best Paper Award. From 2012-2014, he served on the IEEE Trans. on Wireless Communications Editorial Advisory Board. He is an IEEE Senior Member, Secretary of the IEEE Mohawk Valley Section, chair of the IEEE Mohawk Valley Signal Processing Society, and member of Tau Beta Pi and Eta Kappa Nu Engineering Honor Societies.

Find out more »
Free

CESG Teleseminar: Improving Cyber Security through Cyber Insurance and Data Analytics

March 3 @ 4:10 pm - 5:10 pm
WEB, Room 236-C,
Wisenbaker Engineering Building
+ Google Map

Parinaz Naghizadeh, University of Michigan Attempts to improve the state of cyber security have been on the rise over the past decade. In addition to enhancing existing software and infrastructure, there is a parallel need for incentivizing the adoption of these improved security practices by end users and organizations. My research aims to design such incentive mechanisms, and to leverage advances in data analytics for informed cyber-policy design. In this talk, I will first discuss the design of cyber insurance contracts, with an emphasis on users’ unobservable security decisions (moral hazard) and their interdependence in security. I will demonstrate the role of cyber insurance in instilling commitment towards improved cyber security by leveraging users’ interdependence. In addition, I will describe how predictive analytics based on machine learning techniques can be used as a tool for improving the design of these cyber-insurance contracts, and also for regulating security information sharing agreements. Further, I will present a game-theoretic framework for understanding individual users’ decisions towards security investments, and in particular, the effects of the network structure on the outcomes of their interactions. I will discuss how our findings extend several existing results in the literature, as well as their applications in other domains, including the study of spread of research and innovation, financial markets, and environmental pollution reduction policies. Bio: Parinaz Naghizadeh is a postdoctoral research fellow in EECS at the University of Michigan. Her research interests include cyber security, game theory, network economics, optimization, and data analytics. She received her Ph.D. in electrical engineering from the University of Michigan in 2016, M.Sc. degrees in electrical engineering and mathematics, both from the University of Michigan, in 2013 and 2014, respectively, and a B.Sc. in electrical engineering from Sharif University of Technology, Iran, in 2010. She was a recipient of the Barbour scholarship in the 2014-15 academic year.

Find out more »
Free

CESG Teleseminar: Mixed Centralized/Decentralized Decision Protocols for Multi-Agent Systems

March 2 @ 2:30 pm - 4:00 pm

Abstract Multi-agent systems arise in diverse fields, including power systems, robotics, cyber-physical systems, and the Internet of Things. Coordinating these systems is often done using decentralized interactions, in which each agent only communicates with a small number of others. Decentralized algorithms offer several benefits, though they may have difficulty accommodating some performance demands, such as user privacy requirements. Toward addressing such challenges, I will present recent work on mixed centralized/decentralized decision protocols for multi-agent systems. Motivated by the availability of cloud computing, a centralized cloud computer is added to networks of agents in order to gather global information, perform centralized computations, and broadcast the results. As this happens, the agents continue to execute a decentralized behavior. The centralized nature of the cloud means it will be slower than the agents, though its slow, occasional transmissions do indeed enable multi-agent systems to handle various practical challenges. To this end, I will present mixed centralized/decentralized coordination algorithms that tolerate asynchronous information sharing and user privacy requirements, while still enabling strong theoretical guarantees of performance. In the asynchronous case, I will present an algorithm that allows each agent to perform useful work even if the agents have conflicting information about the network. For privacy, the framework of differential privacy is used, giving rise to a novel stochastic optimization algorithm. These algorithms draw from primal-dual optimization techniques and the theory of stochastic variational inequalities, and solve coordination tasks that are stated as convex optimization problems. The end result is a flexible coordination framework that tolerates an array of practical challenges, all while solving constrained coordination problems for teams of agents, regardless of whether an agent is a robot, a self-driving car, or any other physical entity. In addition to theoretical results, I will present robotic implementations of this work to demonstrate its applicability in practice. Bio Matthew Hale is a Ph.D. Candidate in Electrical and Computer Engineering at the Georgia Institute of Technology. In 2012, he received his B.S.E. in Electrical Engineering from the University of Pennsylvania, where he was a member of the GRASP Lab. He received his M.S. in Electrical and Computer Engineering from Georgia Tech in 2015, and was awarded the Colonel Oscar P. Cleaver Outstanding Graduate Student Award by the same department in 2013. His research interests include optimization and control for multi-agent systems, differential privacy, and hybrid systems. His work applies methods from these areas to cyber-physical systems and teams of robots.

Find out more »
+ iCal Import Listed Events