Title: Empowering Devices

Abstract

When I think back, at a device and its accompanying service that truly gave people an unexpected new ability I can¹t help but think back to the invention of the telephone itself. When I think about what people had to do prior to that to talk to their loved ones or friends, they basically had to get on a boat, a horse, or walk for miles. Not only was this long, it was dangerous. The telephone made it possible for people to talk to other people from the comfort of their home. Today we can not only hear each other but see each other too. Whether your across the street or around the globe. In this talk I will review a research program at AT&T Labs - Research called Empowering Devices. The vision of this research work is to create novel devices and companion services that will exploit the wireless and wired network to stimulate demand, while exciting people with unexpected new abilities.

Biography

Dr. Abella is Executive Director of the Innovative Devices and Services Research Department at AT&T Labs, where she manages a group of multi-disciplinary technical staff specializing in human-computer interaction, mobile services, SIP/VoIP technology, ubiquitous computing, and emerging devices.

In May 2013, Dr. Abella will receive Columbia University’s Medal of Excellence, an award given each year to an alumnus or alumna, under 45 years of age, whose record in scholarship, public service, or professional life is outstanding. This is the first time since 1929 -- when the award was first given -- that Columbia has awarded the medal to an engineer. In 2011, she was selected by President Obama to be on his Presidential Advisory Commission for Educational Excellence for Hispanics. Also in 2011, she was inducted into the prestigious WITI(Women in Technology International) Hall of Fame. In 2010, she was honored as one of the Top Five Women of the Year by Hispanic Business Magazine. She is also a member of the elite group of AT&T Science and Technology Medal award winners and recipient of the Hispanic Engineers National Achievement Award for Outstanding Technical Achievement.

Besides her technical contributions, Dr. Abella has been a strong advocate infostering the development of minorities and women in science and engineering. As Executive Vice President for the Young Science Achievers program,a program she has been involved with for 9 years, she has worked tirelessly to bring an interest and excitement in science and engineering to high school aged women and minority students through a program of mentoring and scientific achievement.

Following completion of her Ph.D. in Computer Science from Columbia University in 1995, Dr. Abella joined AT&T Bell Laboratories in Murray Hill, NJ where she began her work in the problem of natural language understanding and dialog management. In addition to her doctoral work, Dr. Abella received a MS from Columbia and BS from NYU, both in Computer Science. She lives in Morristown, New Jersey, with her husband Alex and their son Mark.

Title: "Studying the environment as a complex system: breaking the barrier of traditional assumptions"

Abstract

Traditional signal processing approaches are often built on well-known assumptions such as linearity, time-invariance, and incoherence among the signals involved. In reality, many of these assumptions are violated in physical environments and therefore, conventional techniques fail to solve signal processing issues effectively in real-world systems. For example, a classic challenge to quantifying the environmental impact of a major oil spill such as the 2010 Macondo well spill in the Gulf of Mexico lies in distinguishing its fingerprint against closely correlated natural seeps and other reservoirs in the area. Multi-user detection in the presence of high correlation and amplifier saturation, tracking the time- varying delay spread of shallow water acoustic channels, sonar detection against severe clutter, among many other applications also face similar challenges.

My talk will focus on some of these problems encountered in challenging physical environments, such as the ocean, as they relate my research in adaptive signal processing, sparse optimization and signal separation techniques. While the techniques I will discuss have been developed with target applications in mind, the scope of signal processing innovations go beyond specific applications to the broader paradigm of other well-known problems with similar issues. In particular, I will present my vision on how breaking the barrier of traditional assumptions and constraints can enable the scientific study of the environment as a diverse, complex and dynamic system, with the end goal of enhancing better balance between anthropogenic activities and environmental health and sustenance.

Biography

Dr. Ananya Sen Gupta got her MS (Aug. 2001) and PhD (Dec. 2006) in Electrical Engineering from the University of Illinois at Urbana-Champaign. Her broad area of expertise is signal processing, particularly interference mitigation, channel estimation and sparse optimization techniques. Throughout her research career Dr. Sen Gupta has been interested in seeking novel solutions for signal processing challenges commonly encountered in real environments. Her Master's research was in dynamic MRI imaging under the supervision of Dr. Zh-Pei Liang, and her doctoral research was in non-linear multi-user interference suppression under the supervision of Dr. Andrew Singer. Aug 2008-2012 she has been a postdoctoral researcher at the Woods Hole Oceanographic Institution under the supervision of Dr. James Preisig, working in shallow water acoustic channel tracking using adaptive sparse optimization techniques. As her continuing position at WHOI, she has been actively collaborating with Dr. Christopher Reddy to develop novel informational methods that provide collective forensic interpretation of petroleum samples found in the ocean using comprehensive two-dimensional gas chromatography. More recently, she has joined the University of Iowa, Department of Electrical and Computer Engineering, as tenure-track faculty. Her active research interests involve signal processing challenges commonly encountered in complex environments, with particular focus on fingerprinting and sparse optimization across a variety of applications.

Title: "Inter-temporal Price Discrimination: Structure and Computation of Optimal Policies"

Abstract

We consider the question of how should a firm optimally set a sequence of prices in order to maximize its long-term average revenue given a continuous flow of strategic customers. In particular, customers arrive over time, are strategic in timing their purchases and are heterogeneous along two dimensions: their valuation for the firm's product and their willingness to wait before purchasing or leaving. The customers' patience and valuation may be correlated in an arbitrary fashion. For this general formulation, we prove that the firm may restrict attention to short cyclic pricing policies, which have length twice the maximum willingness to wait of the customer population. We further establish results on the subpotimality of monotone pricing policies in general, and illustrate the structure of optimal policies. These are, in a typical scenario, characterized by nested sales, where the firm offers partial discounts throughout each cycle, offers a significant discount halfway through the cycle, with the largest discount offered at the end of the cycle. From a computational perspective, we exploit the structure of the underlying problem to develop a novel dynamic programming formulation for the problem that computes an optimal pricing policy in polynomial time (in the maximum willingness-to-wait). We further establish a form of equivalence between the problem of pricing for a stream of heterogeneous strategic customers and pricing for a pool of heterogeneous customers who may stockpile units of the product. Joint work with Omar Besbes (Columbia).

The paper is available here:
http://pages.stern.nyu.edu/~ilobel/IntertemporalPriceDiscrimination.pdf

Biography

Dr. Ilan Lobel is an Assistant Professor of Information, Operations and Management Sciences at New York University's Stern School of Business. Prior to joining NYU Stern in 2010, he was a post-doctoral researcher at the Microsoft Research New England Lab. He received his Ph.D. in Operations Research from the Massachusetts Institute of Technology in 2009 and his B.Sc. in Electrical Engineering from the Pontificia Universidade Catolica of Rio de Janeiro in 2004. Professor Lobel's research focuses on the operations of Internet-based businesses, including issues such as pricing, learning and contract design in online, dynamic and networked markets. His work focuses on application domains such as online advertisement, cloud computing and social networks.

Abstract

Lithography is used to pattern semiconductor devices such as memories and processors. Water immersion lithography using the ArF excimer laser source at 193nm is the key manufacturing tool today. EUV lithography is developed to continue the shrink of integrated circuit features below 20nm half pitch.

Laser produced plasma sources operating at 13.5nm are used with all-reflective optics to expose resists. This presentation will describe the status of EUVL and the remaining challenges to bring it into high volume manufacturing by 2015.

Biography

Bill Arnold is for 15 years the chief scientist of ASML, the world's leading supplier of lithography equipment for semiconductor manufacturing. He is also Sr. Vice President of the ASML Technology Development Center. Before that he was 18 years with Advanced Micro Devices, a leading manufacturer of microprocessors. He had many jobs over time at AMD, the last being Director of Advanced Process Development and Sr. Fellow. In 2013, Mr. Arnold is the President of SPIE, the International Society for Optics and Photonics.

Title: Transforming computing algorithms and paradigms in HPC to enable more science out of our day-to-day simulations

Abstract

My research group at the University of Delaware is engaged in interdisciplinary research with scientists and engineers in fields such as chemistry, chemical engineering, pharmaceutical sciences, seismology, and mathematics. The group's mission is to increase scientific discovery by transforming the way scientific applications deploy high performance computing (HPC) including distributed and multi-core systems. Several research vignettes will be presented that show how we have been successful in addressing challenges in science such as classifying protein-ligand binding geometries of billions of ligands in molecular docking in linear time, performing accurate simulations of fully atomistic macromolecular systems at meso-length and time scales, and improving numerical reproducibility and stability of MD simulations on multi-core platforms.

Biography

Michela Taufer joined the University of Delaware in 2007 as an assistant professor. She was promoted to associate professor with tenure in 2012. In January 2013 she was named the David L. and Beverly J.C. Mills Chair of Computer and Information Sciences. She earned her master's degrees in Computer Engineering from the University of Padova (Italy) and her doctoral degree in Computer Science from the Swiss Federal Institute of Technology (Switzerland). Taufer completed her postdoctoral studies as a La Jolla Interfaces in Sciences Fellow at the Center for Theoretical Biological Physics (CTBP), at the University of California, San Diego. Taufer's research focuses on efficient computational algorithms and adaptive scheduling policies for scientific computing on HPC platforms including multi-core, cloud, and volunteer computing.

Sparsity and Interpretability in Predictive Multivariate Analysis of fMRI Data

Abstract

One of the central topics in statistical analysis of fMRI data is discovering brain areas relevant to a given stimuli or mental state. Herein, we focus on predictive accuracy as a better relevance measure than the traditional voxel activations based on univariate voxel correlations with stimulus, since the latter approach ignores potentially important multivariate voxel interactions.

Since an exhaustive search over all subsets of voxels is intractable, sparse (l1-regularized) regression is a popular alternative for learning predictive models simultaneously with selection of predictive subsets of voxels, since l1 constraint on the regression model parameters tends to find solutions with relatively small number of nonzeros. However, multiple well-predicting sparse solutions may exist when the variables are highly correlated, as it is the case in fMRI data. Thus, while the brain areas included into a highly predictive sparse solution are clearly relevant to the task, it is not clear how unique such solutions are, i.e. how much information about the task is still contained in the rest of the brain.

This leads to the following questions: should one expect a sharp boundary between task-relevant and task-irrelevant brain areas, or rather a widespread distribution of task-relevant information across the whole brain? How does the task-related information distribution depend on the properties of the task? Herein, empirical exploration of such boundary is performed using the Elastic Net regression when predicting several stimuli and/or behavior variables from fMRI, including pain perception, visual stimulus rating, and several behavioral measures recorded during videogame playing (PBAIC 2007 data). Interestingly, for most of the tasks, no clear separation between relevant and irrelevant areas is observed, with the only exception of a relatively simple auditory task.

Thus, we hypothesize that complex tasks tend to involve most of the brain rather than just specific areas, which points in the direction of ``holographic'' information representation for such tasks, while simpler tasks yield more clear separation between relevant and irrelevant areas. These observations suggest a novel methodological approach that goes beyond traditional voxel activation maps and involves a more comprehensive evaluation of information spread in the brain.

Biography

Irina Rish , is a Research Staff Member (RSM) at IBM T.J. Watson Research Center. She received her MS in Applied Mathematics from Moscow Gubkin Institute, Russia, and PhD in Computer Science from the University of California, Irvine. Irina Rish research interests are in the areas of probabilistic inference, machine learning, and information theory. Particularly, she has done work on approximate inference in graphical models, information-theoretic experiment design and active learning, with applications are in the area of autonomic computing - automated management of complex distributed systems, which includes various diagnosis, prediction and online decision-making problems. Irina Rish’s current research is in the area of machine-learning applications to computational biology and neuroscience, with a particular focus on statistical analysis of brain imaging data such as fMRI.

Abstract

The viability of photovoltaic energy has increased tremendously in the last decade as new manufacturing capacity (especially in China) has driven PV module costs to levels never thought possible. However, there remain challenges in achieving true grid-cost-parity for PV compared to other sources such as coal and natural gas. These challenges include balance of system components and system integration costs, both of which are now major focuses of the DOE. This talk will discuss PV systems from a general perspective, targeting audience members who have less exposure to the PV area. The basic configurations of PV systems will be presented from the cell, module, string, array, and power plant level. We will then drill down into the different architectures for the power management system including central inverter, micro-inverter, and other emerging architectures. New classes of multi- level DC-DC converter ‘partial-power processing’ circuits will be presented as next-generation solutions.

Biography

Dr. Stauth received his MS and PhD degrees at University of California, Berkeley in 2006 and 2008 respectively. At U.C. Berkeley Dr. Stauth was affiliated with the Berkeley Wireless Research Center (BWRC) and the Power Electronics Research Group, studying under Prof. Seth Sanders. His research focused on high frequency power electronics and RF integrated circuits. In 2008 he co-founded QVSense, Inc. and operated as CTO until QVSense was acquired by Solar Semiconductor. He remained as Director of Hardware Engineering at Solar Semiconductor until 2011. Since 2011 he has been Assistant Professor at Dartmouth College.

Abstract:

While machine vision research has continued to improve multi-fold over the past few decades, it still significantly falls short of the abilities and efficiencies of the primate visual cortex system. The primate brain is especially superior as pertains to comprehending and interacting with complex natural environments. In energy-efficiencies, the brain is estimated to be around four orders better than current machine vision solutions. While there is much consensus on the superiority of biological vision systems over machine systems on most vision tasks, the approaches leading to the better efficiencies and flexibility akin to the visual cortex are still widely debated. In this talk, I will highlight recent efforts at architecting customized digital hardware systems using neuromorphic algorithms as one successful approach to achieving better energy efficiencies. I will also provide insight to how emerging devices are influencing novel implementations of such hardware systems.

Biography:

Vijaykrishnan Narayanan is a Professor of Computer Science and Engineering and Electrical Engineering at The Pennsylvania State University. His research and teaching interests include embedded systems, computer architecture, system design using emerging device technologies and power-aware computing. He has deep interests in cross-disciplinary advances and has led and participated in such projects. He is the deputy editor-in-chief of IEEE TCAD and served as the editor-in-chief for ACM Journal of Emerging Technologies in Computing Systems. He has won several awards including the 2012 ASPDAC Ten-year retrospective Most influential paper, 2012 Penn State Alumni Society Premier Research Award and 2010 Outstanding Alumnus Award from SVCE, India. He is a fellow of IEEE. He has worked with several outstanding students who are currently in industry and academia throughout the world.

Photovoltaic Energy: A Circuits and Systems Perspective

Abstract:

The viability of photovoltaic energy has increased tremendously in the last decade as new manufacturing capacity (especially in China) has driven PV module costs to levels never thought possible. However, there remain challenges in achieving true grid-cost-parity for PV compared to other sources such as coal and natural gas. These challenges include balance of system components and system integration costs, both of which are now major focuses of the DOE. This talk will discuss PV systems from a general perspective, targeting audience members who have less exposure to the PV area. The basic configurations of PV systems will be presented from the cell, module, string, array, and power plant level. We will then drill down into the different architectures for the power management system including central inverter, micro-inverter, and other emerging architectures. New classes of multi-level DC-DC converter ‘partial-power processing’ circuits will be presented as next-generation solutions.

Biography:

Dr. Stauth received his MS and PhD degrees at University of California, Berkeley in 2006 and 2008 respectively. At U.C. Berkeley Dr. Stauth was affiliated with the Berkeley Wireless Research Center (BWRC) and the Power Electronics Research Group, studying under Prof. Seth Sanders. His research focused on high frequency power electronics and RF integrated circuits. In 2008 he co-founded QVSense, Inc. and operated as CTO until QVSense was acquired by Solar Semiconductor. He remained as Director of Hardware Engineering at Solar Semiconductor until 2011. Since 2011 he has been Assistant Professor at Dartmouth College.

Abstract

Sparsity has emerged as a powerful modeling concept in a wide range of signal processing and learning problems: Many high-dimensional objects admit a sparse representation in a suitable transform basis and the recent field of compressed sensing has provided efficient methods to exploit this sparse structure in estimation, dimensionality reduction and feature extraction. However, sparsity is generally not the only aspect of many practical inverse problems. This talk presents a powerful new class of algorithms called generalized approximate message passing (GAMP) that significantly extends the compressed sensing framework. The GAMP algorithm use Gaussian approximations of loopy belief propagation to reduce estimation problems on complex interconnected systems to smaller, local problems associated with the individual system components. For compressed sensing estimation, the GAMP method can incorporate arbitrary priors and nonlinearities, perform joint estimation of latent variables, and can exploit large classes of complex statistical relationships between variables. Moreover, for certain inverse problems with large random transforms, the GAMP method admits remarkably precise asymptotic performance characterizations with testable conditions for optimality -- even for problems that are non-convex and nonlinear. Applications are demonstrated in challenging identification problems in neuroscience and in optimization problems in wireless interference coordination. Joint work with Alyson Fletcher (UCSC), Phil Schniter (Ohio State), Ulugbek Kamilov (EPFL), Vivek K Goyal (MIT) and Lav Varshney (IBM).

Biography

Dr. Sundeep Rangan received the B.A.Sc. at the University of Waterloo, Canada and the M.Sc. and Ph.D. at the University of California, Berkeley, all in Electrical Engineering. He has held postdoctoral appointments at the University of Michigan, Ann Arbor and Bell Labs. In 2000, he co-founded (with four others) Flarion Technologies, a spin off of Bell Labs, that developed Flash OFDM, one of the first cellular OFDM data systems, and precursor to 4G cellular standards such as LTE and WiMax. Flarion grew to over 150 employees with trials worldwide. In 2006, Flarion was acquired by Qualcomm Technologies where Dr. Rangan was a Director of Engineering involved in OFDM infrastructure products. He joined the ECE department at NYU-Poly in 2010. His research interests are in wireless communications, signal processing, information theory and control theory.

Title:
Scientific Challenges in Information Retrieval from Earth Observation (EO) Imagery

Abstract:

The ultimate goal of any Earth Observation (EO) system is to provide understanding. However, this understanding will often require expertise and/or data sources from globally distributed resources and will present some unique challenges for the local remote sensing analyst. To address these challenges, it is incumbent upon the global community to evolve and sustain a global observation network. These observations serve as the foundation for the models that are used to describe Earth processes. As this observational data accumulates in global archives new opportunities become available for knowledge discovery about the Earth system. However, access to these observational data is optimized for the science teams for whom the instruments were launched and access by operational users may be problematic. This presentation will lay out some of the Grand Challenges for those engineers and scientists involved in pattern recognition in the Earth remote sensing arena. It describes the problem space for making local decisions and introduces the concept of contextual remote sensing. Answers will require addressing the multi-dimensional aspects of the EO imagery – hyper-spatial, hyper-spectral, hyper-temporal, and hyper sensors.

Biography:

Nicolas H. Younan is currently the Department Head and James Worth Bagley Chair of Electrical and Computer Engineering at Mississippi State University. He received the B.S. and M.S. degrees from Mississippi State University, in 1982 and 1984, respectively, and the Ph.D. degree from Ohio University in 1988. Dr. Younan’s research interests include signal processing and pattern recognition. He has been involved in the development of advanced image processing and pattern recognition algorithms for remote sensing applications, image/data fusion, feature extraction and classification, automatic target recognition/identification, and image
information mining.

Dr. Younan has published over 200 papers in refereed journals and conference proceedings. He has served as the General Chair and Editor for the 4th IASTED International Conference on Signal and Image Processing, Co-Editor for the 3rd International Workshop on the Analysis of Multi-Temporal Remote Sensing Images, Guest Editor, Pattern Recognition Letters and IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (JSTARS), and Co-Chair, 4th and 5th Workshop on Pattern Recognition for Remote sensing (PRSS). He is a senior member of IEEE and a member of the IEEE Geoscience and Remote Sensing society, serving on two technical committees: Data Fusion, and Data Archive and Distribution. He also served as the Vice Chair of the International Association on Pattern Recognition (IAPR) Technical Committee 7 on Remote Sensing (2008-2010).

ABSTRACT

It has been more than a decade since the Blue Gene project started in IBM Research. Blue Gene/Q, the latest and third generation of Blue Gene, is again at the top of the Supercomputer TOP500 and Green500 lists.

This talk will highlight the evolution of the Blue Gene technologies and its revolutionary impact on Sciences. I will also discuss the new challenges and directions of Supercomputing beyond Blue Gene.

BIOGRAPHY

Dr. Kyung Dong Ryu is Manager and Research Project Leader of the Exascale System Software Research Group at the IBM T.J. Watson Research Center. Since 2004, he has led several research projects in High Performance Computing, Cloud Computing, Resilient HA Cluster, End-to-End System Performance Optimization. He had also led the scalable system management project for datacenters, advancing IBM’s systems management products, for which he received an Outstanding Technical Achievement Award from IBM. Before joining IBM Research, he was an Assistant Professor at the Arizona State University (2001-2004), where his research focused on PC Grid and Operating Systems for High Performance Computing. As a faculty member, he was awarded NSF grants and led several industrial research projects funded by the Research Consortium with Intel and Motorola. He published over 40 papers in premier international conferences and journals, including SC, ICS, IPDPS, ICDCS, TPDS, Eurosys, and ISCA. He received M.S. and Ph.D. in 1998 and 2001, respectively, from the University of Maryland, College Park, in Computer Science.

Abstract

The dramatic increases in demands from multimedia applications are putting an enormous strain on current cellular system infrastructure. Hence, we are witnessing significant research and development efforts on 4G multi-channel wireless cellular systems (e.g., LTE and WiMax) that target new ways to achieve higher data rates, lower latencies, and a much better user experience. An important requirement for achieving these goals is to design efficient scheduling policies that can simultaneously provide high throughput and low delay. In these multi-channel systems, such as OFDM, the Transmission Time Interval (TTI), within which the scheduling decisions need to be made, is typically on the order of a few milliseconds. On the other hand, there are hundreds of orthogonal channels that can be allocated to different users. Hence, many decisions have to be made within a short scheduling cycle and it is critical that scheduling policies must be of low complexity. In this talk, we will present a unifying framework for designing low-complexity scheduling policies in the downlink of multi-channel (e.g., OFDM-based) wireless networks that can provide optimal performance in terms of both throughput and delay. We first develop new easy-to verify sufficient conditions for rate-function delay-optimality in the many-channel many-user asymptotic regime. and for throughput-optimality in non-asymptotic settings.

The sufficient conditions enable us to prove rate-function delay optimality for a class of Oldest Packets First (OPF) policies and throughput optimality for a large class of Maximum Weight in the Fluid limit (MWF) policies, respectively. While a recently developed scheduling policy is both throughput-optimal and rate-function delay-optimal, it has a very high complexity of O(n^5), where n is the number of channels or users, which makes it impractical. By intelligently combining policies from the classes of OPF policies and MWF policies, we design hybrid policies that have a low complexity of O(n^{2.5} log n), without losing either throughput-optimality or rate-function delay-optimality. We further develop two simpler greedy policies that are both throughput-optimal and have a positive rate-function. We show through simulations that empirically these simpler mechanisms have near-optimal value of rate-function in various scenarios. Finally, we propose a class of throughput-optimal policies with even lower complexity that allows an explicit trade-off between complexity and delay performance.

Biography

Ness Shroff received his Ph.D. degree in Electrical Engineering from Columbia University in 1994. He joined Purdue university immediately thereafter as an Assistant Professor in the school of ECE. At Purdue, he became Full Professor of ECE in 2003 and director of CWSA in 2004, a university-wide center on wireless systems and applications. In July 2007, he joined The Ohio State University, where he holds the Ohio Eminent Scholar endowed chair professorship in Networking and Communications, in the departments of ECE and CSE. Since 2009, he also serves as a Guest Chaired professor of Wireless Communications at Tsinghua University, Beijing, China. His research interests span the areas of communication, social, and cyberphysical networks. He is especially interested in fundamental problems in the design, control, performance, pricing, and security of these networks. Dr. Shroff is a past editor for IEEE/ACM Trans. on Networking and the IEEE Communication Letters. He currently serves on the editorial board of the Computer Networks Journal, IEEE Network Magazine, and the Networking Science journal. He has chaired various conferences and workshops, and co-organized workshops for the NSF to chart the future of communication networks. Dr. Shroff is a Fellow of the IEEE and an NSF CAREER awardee. He has received numerous best paper awards for his research, e.g., at IEEE INFOCOM 2008, IEEE INFOCOM 2006, IEEE IWQoS 2006, Journal of Communication and Networking 2005, Computer Networks 2003, and one of two runner-up papers at IEEE INFOCOM 2005.

Abstract:

Many scientifically important questions require the efficient use of high-performance and distributed computing in order to provide answers with the accuracy needed at the length and time-scales required. We begin by analyzing how and why it has been necessary to develop "effective abstractions" in order to successfully utilize production high-performance distributed cyber-infrastructure, such as NSF's TeraGrid/XSEDE. For example, pilot-jobs are arguably one of the most widely-used distributed computing abstractions, and have been shown to support scalable and dynamic utilization of distributed resource.

However, there does not exist a well-defined, unifying conceptual model of pilot-jobs which can be used to define, compare and reason across different implementations of pilot-jobs; this presents a barrier to extensibility and interoperability. We introduce the P* Model the first known conceptual model of pilot-jobs, validate its implementation via the Pilot-API -- by concurrently using multiple distinct pilot-job frameworks on distinct production distributed cyber-infrastructures, and propose extensions of the P* Model to data.

We will discuss the application of the pilot-abstraction to support the infrastructural and algorithmic requirements of several Grand Challenge problems facing the Computational Biology community, for example data-analytics for next-generation gene sequencing, enhanced sampling molecular algorithms, and in-silico personalized and predictive health-care.

For Graduate students looking for research opportunities, we will conclude by providing a brief overview of the Radical group and some research topics currently of interest.

Biography:

Prof. Shantenu Jha is an Assistant Professor at Rutgers University, a member of the Graduate Faculty in the School of Informatics at the University of Edinburgh (UK), and a Visiting Scientist at University College London. Before moving to Rutgers, he was the lead for Cyberinfrastructure Research and Development at the CCT at Louisiana State University. His research interests lie at the triple point of Computer Science, Cyberinfrastructure Development and Computational Science.

Shantenu is the lead investigator of the SAGA project ( http://www.saga-project.org), which is a community standard and is part of the official middleware/software stack of most major Production Distributed Cyberinfrastructure -- such as US NSF's XSEDE and the European Grid Infrastructure. His research has been funded by multiple NSF awards, US National Institute for Health (NIH) as well as the UK EPSRC (OMII-UK project and Research theme at the e-Science Institute). Jha is the author of approximately 60 publications; he has won some of the most prestigious awards in high-performance computing at ACM/IEEE Supercomputing and the International Supercomputing Series.

Abstract:

Pressure on financial models and time to market create a relationship between speed of computation and business success. This talk explains how the application of dataflow techniques by JP Morgan has achieved cutting edge performance, whilst also helping to improve trading strategy, decision making and risk management.

Biography:

Stephen is currently head of the Applied Analytics group within the investment banking division of JP Morgan. The group is responsible for accelerating mathematical models across asset classes for trading and risk management using dataflow computational techniques. Prior to joining JP Morgan Stephen spent lengthy periods at Deutsche Bank, Credit Suisse, Barclays and UBS. Prior to entering investment banking he also spent 5 years as a university lecturer teaching mathematical economics, banking, finance and monetary theory. Stephen holds a PhD in mathematical finance from Cass Business School in London.

Abstract

Wireless devices and sensors are increasingly ubiquitous in all aspects of life. As a result, researchers have worked tirelessly to provide more functionality and ever higher data rates to the devices. Researchers are challenged to use energy more efficiently, due to finite battery capacity and increasingly as everyone is asked to reduce their demands from the electric grid. In this talk I will address the challenge of using energy more efficiently in wireless communications systems by leveraging linearization around CMOS switching amplifiers. These switching amplifier topologies provide means to increase output power, efficiency and integratability of the PA with the rest of the radio circuitry, a major stumbling block in the quest for the RF system-on-a-chip (SOC).

I will summarize why switching amplifiers can outperform their linear counterparts and offer potential for tunability and reconfigurability for software defined radio (SDR) applications. Next I will motivate linearization methods that allow switching PAs to be used with non-constant envelope modulation, including pulse-width and -position modulation (PWPM) and envelope elimination and restoration (EER). The switched-capacitor PA, a topology that utilizes switched capacitors to enable a significant improvement in average efficiency and linearity utilizing a combination of data converter and power amplifier techniques will be introduced. It represents an exciting path towards SDR ready CMOS power amplifiers. I will conclude the talk with a few interesting research directions in energy efficient RF CMOS circuit design.

Biography

Dr. Walling received the B.S. degree from the University of South Florida, Tampa, in 2000, and the M.S. and Ph. D. degrees from the University of Washington, Seattle, in 2005 and 2008, respectively. Prior to starting his graduate education he was employed at Motorola, Plantation, FL working in cellular handset development. He interned for Intel, Hillsboro from 2006-2007, where he worked on highly-digital transmitter architectures and CMOS power amplifiers and continued this research while a Postdoctoral Research Associate with the University of Washington. He is currently an assistant professor in the electrical and computer engineering department at Rutgers, The State University of New Jersey.

His current research interests include low-power wireless circuits, energy scavenging, high-efficiency transmitter architectures and CMOS power amplifier design for software defined radio. Dr. Walling has authored over 30 articles in peer reviewed journals and refereed conferences. He received the Yang Award for outstanding graduate research from the University of Washington, Department of Electrical Engineering in 2008, an Intel Predoctoral Fellowship in 2007-2008, and the Analog Devices Outstanding Student Designer Award in 2006.

Nano-Relay Technology for Energy-Efficient Electronics

Abstract: As the era of traditional Complementary-Metal-Oxide-Semiconductor (CMOS) technology scaling is coming to an end, continual improvements in integrated-circuit (IC) performance and cost per function are becoming difficult to achieve without increasing power density. This necessitates the investigation of alternate device technologies that surmount the fundamental CMOS energy-efficiency limit and hence enable ultra-low-power ICs. To that end, a nano-electro-mechanical (NEM) relay technology is promising, because of its immeasurably low off-state leakage current and abrupt turn-on behavior, which provide for zero static power consumption and potentially very low dynamic power consumption.

In this talk, I will discuss recent research efforts in NEM relay technology, from both device- and circuit-level perspectives, which led to the successful demonstration of relay-based digital IC building blocks. In addition, I will discuss multi-input relay devices that can lead to smarter design and compact implementation of zero-leakage digital integrated circuits.

Biography: Jaeseok Jeon received the B.A.Sc. degree with first-class honours in electronics engineering from Simon Fraser University, Canada, in 2007 and the Ph.D. degree in electrical engineering from the University of California, Berkeley, in 2011. In 2011, he joined the Rutgers, the State University of New Jersey, as an assistant professor of Electrical and Computer Engineering.

In 2006, he was an electronics designer at Kodak Graphic Communications Canada Company, and he was awarded the 2006 NSERC-USRA award. He was a co-recipient of the 2011 ISSCC Jack Raper Award for Outstanding Technology Directions. His research interests include nano-electro-mechanical relay devices and technology for energy-efficient electronic systems and neural devices and circuits for efficient design of neuromorphic systems.

“Quantitative Data Convergence: Fusing radiology, pathology, omics data for predicting disease aggressiveness and patient outcome"

Abstract: Traditional biology generally looks at only a few aspects of an organism at a time and attempts to molecularly dissect diseases and study them part by part with the hope that the sum of knowledge of parts would help explain the operation of the whole. Rarely has this been a successful strategy to understand the causes and cures for complex diseases. The motivation for a systems based approach to disease understanding aims to understand how large numbers of interrelated health variables, gene expression profiling, its cellular architecture and microenvironment, as seen in its histological image features, its 3 dimensional tissue architecture and vascularization, as seen in dynamic contrast enhanced (DCE) MRI, and its metabolic features, as seen by Magnetic Resonance Spectroscopy (MRS) or Positron Emission Tomography (PET), result in emergence of definable phenotypes. At the Laboratory for Computational Imaging and Bioinformatics, we have been developing computerized knowledge alignment, representation, and fusion tools for integrating and correlating heterogeneous biological data spanning different spatial and temporal scales, modalities, and functionalities. The long term research objectives are to explore via efficient computational and pattern recognition methods, the existence and correspondence of biological patterns across heterogeneous data scales and modalities. An understanding of the interplays of different hierarchies of biological information from proteins, tissue, metabolites, and imaging will provide conceptual insights and practical innovations that will profoundly transform people’s lives.

ABSTRACT

Rank aggregation is the problem of combining multiple candidate rankings into one list that best reflects the candidates' standing as a whole. Rank aggregation has many applications, in fields as diverse as bioinformatics, coding theory and social sciences.

Mathematically, the rank aggregation problem can be formulated as finding a permutation that represents the ``centroid'' of a set of permutations - i.e., a permutation that minimizes a given average distance function from the given set of permutations. The main issue arising in such aggregation problems is to identify distance functions that are scalable, flexible and easy to compute.
So far, no solutions that have these desirable properties were proposed.

We introduce a new class of cost-constrained permutation metrics that can be approximated within a constant in polynomial time. The cost functions are based on average transposition distances and for costs that have a tree-metric form, the presented algorithms are exact. To prove the optimality of the distance computation procedure, we use Menger's theorem and graphical representations of permutations.

We conclude the talk by describing a number of applications of the novel distance metric in ``collaborative rank aggregation'' and coding for multilevel flash memories.
__________________________________________________________________________________________

BIOGRAPHY

Olgica Milenkovic received a MSc degree in mathematics and PhD degree in electrical engineering from the University of Michigan, Ann Arbor, in 2001 and 2002, respectively. From 2002 until 2006 she was with the faculty of University of Colorado, Boulder. In 2006, she was a visiting professor at the University of California, San Diego Center for Information Theory and Application. In 2007, she joined University of Illinois, Urbana-Champaign were she currently holds the position of associate professor. Her research interests are in algorithms, bioinformatics, coding theory, combinatorics and signal processing.

Olgica Milenkovic is a recipient of the NSF Career Award, the DARPA Young Faculty Award and a number of best conference paper awards. She currently serves as Associate editor in the Transactions on Signal Processing and the Transactions on Information Theory.

Abstract

Thirty years ago, most large companies had pension plans that promised retirees a "defined" pension-- usually about 1% of a person's final salary for each year worked. Someone with 40 years of service would therefore retire with about 40% of their final salary, plus another 20% or so from Social security. Some pensions even provided cost-of-living adjustments. This allowed a modest lifestyle in retirement or, with some savings, a comfortable and attractive lifestyle. In particular, these plans guaranteed income for life, at a time when people were retiring earlier and living longer.

From the employer's viewpoint, those "defined benefit" plans represented a substantial liability that could escalate sharply in a time of inflation, and as life expectancies increased. As a result, most employers abandoned these "promises to pay" as too expensive and risky in a competitive world. The "defined benefit" pension is now almost extinct, and a successful life now requires some new skills in planning and investing.

Savings plans like the 401K (and other similar retirement savings accounts) are essential to this task. They defer income taxes on "contributions" (money that is put in), and on growth (money that is earned in the account), until the money is removed. Additionally, some employers also "match" a fraction of the employee's contribution. Compared to saving without a 401K, the tax advantage and employer match can easily triple the "effective pension" that can be achieved (the amount of money a person can eventually take out annually in retirement). Even without an employer match, the "effective pension" can be doubled by the delayed tax alone, so there is a strong incentive to participate in whatever type of plan is available. Those who contribute a reasonable amount to these plans, and who invest the money in their plans intelligently, can look forward to retiring relatively early into lives that are interesting, secure and comfortable. Those who fail to contribute, or who invest foolishly, will work longer than they want to and spend their last years in relative poverty. For the new graduate with a first job, these issues may seem far off and more immediate concerns may seem more important. Frequently, action gets postponed, and with each year of delay the desired result becomes harder to achieve.

In this lecture, we will use a simple planning spreadsheet and some basic financial data to explore the issues of how much to save and how to invest those savings. There is no single answer to these questions, of course. Some will wish to retire earlier or live more expensively in retirement. Some will be willing to take greater risks to achieve their goals. The objective of the lecture is for the student to understand what is possible, and how such lifestyle goals are reflected in the overall plan.

BIOSKETCH

Dick Frenkiel was born more than a half-century ago, in a small town called Brooklyn. He attended Tufts University and Rutgers University, emerging with degrees in Mechanical Engineering and a deceptive aura of competence. Soon after joining Bell Laboratories in 1963, he was mistaken for an electrical engineer of similar height and moved into cellular systems engineering, where for sixteen years he did little serious harm.

During the late 1960's. Dick was an author of AT&T's cellular system proposal to the FCC. After an obligatory "growth experience" at corporate headquarters, during which he acquired several suits, he returned to Bell Labs, where he became head of systems engineering for AMPS, the first cellular system in the US. He invented a method for efficient and low-cost cell-splitting, and served on the EIA committee which defined the first cellular standards to be used in the United States.

For Dick, the joy of seeing AT&T's first cellular service in Chicago was somewhat diminished when the company was torn apart by the government and excluded from the cellular business. While not responsible for this calamity, he exiled himself to the lowly world of consumer electronics. He became head of R&D for AT&T's cordless telephone business unit, and led a team that developed cordless telephones, some of which actually worked. He was also responsible for the first manufacture of these telephones in Singapore and Hong Kong, where he acquired several additional suits at attractive prices.

Dick retired from AT&T in 1993, and spent many happy years at WINLAB, the Wireless Information Networks Laboratory here at Rutgers. In 1999, he served as mayor of Manalapan Township in New Jersey and was not indicted. With Professor Narayan Mandayam he teaches a multi-disciplinary course in business strategy in the wireless industry.

For his work in wireless, Dick has received the National Medal of Technology, the Alexander Graham Bell Medal and the Industrial Research Institute Achievement Award. He is a member of the National Academy of Engineering, a New Jersey Inventor of the Year, and a Fellow of Bell Labs and the IEEE.

Abstract

Silicon photonics offers a low-cost platform for building large-scale photonic devices and circuits, potentially replicating the success of silicon microelectronics in photonics. Photonic crystal nanostructures provide novel physical mechanisms, such as the slow-light effect and superprism effect, to reduce the size and improve the performance of silicon photonic devices. This talk will review our recent work on silicon-compatible waveguides, including slow-light photonic crystal waveguides, for use as on-chip optical delay lines. Particularly, some fundamental physical issues of photonic crystal waveguides such as slow-light loss will be addressed. Optical delay lines have applications in phased array antennas, optical signal processing, and optical modulation. It will be shown that the current slow-light loss levels are sufficient for some applications, whereas challenges remain in other applications. This talk will also review our work on dual racetrack silicon micro-resonators for quadrature amplitude modulation. Potential applications of these device components in optical communications, optical interconnects, laser beam steering, and optical signal processing will be discussed.

Bio sketch

Wei Jiang is an assistant professor in the Department of Electrical and Computer Engineering of Rutgers, the State University of New Jersey. He received his B.S. degree in physics from Nanjing University, Nanjing, China, in 1996, and his M.A. degree in physics and his Ph.D. degree in electrical and computer engineering from the University of Texas, Austin, in 2000 and 2005, respectively. His research interests encompass silicon photonics, photonic crystals, nanophotonics, and their applications in various optoelectronic systems.

Photonic crystals are ultra-compact highly integrated nano scale structures which demonstrated the possibility of generating, manipulating, processing, transmitting and detecting light. However, most photonic crystal devices are “passive” structures without any means of external on-demand control. Adding tunability to the photonic crystals would greatly expand their application areas and enable unforeseen new application areas. MEMS/NEMS technologies are ideally positioned to provide a wide variety of unprecedented radical options of tunability to the passive photonic crystal devices as they can be co-integrated with photonic crystals. We have recently reported MEMS-enabled mechanically tunable and thermally tunable photonic crystals. Various applications and current status of development of such tunable photonic crystals will be discussed.

Speaker Bio:

Dr. Jeong-Bong (JB) Lee received the Ph.D. degree from Georgia Tech, Atlanta, Georgia in 1997. In 2001, he joined the Electrical Engineering Department at UTD where he is now a full professor. His current research interests include MEMS and nanophotonics for biomedical and photonics applications. He received the NSF CAREER AWARD in 2001. He is currently serving as an editorial board member for Micromachines journal. He has served as a program committee member for many international conferences including Transducers 2011 Conference as a member of executive program sub-committee. He also served as a member of external advisory board for the Microsystems division at the Sandia National Laboratories in 2007. He has five U.S. patents, more than 45 journal papers and 138 conference papers published.

"SCION: Scalability, Control, and Isolation on Next-Generation Networks" by Prof. A. Perrig, Carnegie Mellon University

We present the first Internet architecture designed to provide route control, failure isolation, and explicit trust information for end-to-end communications. SCION separates ASes into groups of independent routing sub-planes, called trust domains, which then interconnect to form complete routes. Trust domains provide natural isolation of routing failures and human misconfiguration, give endpoints strong control for both inbound and outbound traffic, provide meaningful and enforceable trust, and enable scalable routing updates with high path freshness. As a result, our architecture provides strong resilience and security properties as an intrinsic consequence of good design principles, avoiding piecemeal add-on protocols as security patches. Meanwhile, SCION only assumes that a few top-tier ISPs in the trust domain are trusted for providing reliable end-to-end communications, thus achieving a small Trusted Computing Base. Both our security analysis and evaluation results show that SCION naturally prevents numerous attacks and provides a high level of resilience, scalability, control, and isolation.

Bio
Adrian Perrig is a Professor in Electrical and Computer Engineering, Engineering and Public Policy, and Computer Science at Carnegie Mellon University. Adrian serves as the technical director for Carnegie Mellon's Cybersecurity Laboratory (CyLab). He earned his Ph.D. degree in Computer Science from Carnegie Mellon University, and spent three years during his Ph.D. degree at the University of California at Berkeley. He received his B.Sc. degree in Computer Engineering from the Swiss Federal Institute of Technology in Lausanne (EPFL). Adrian's research revolves around building secure systems and includes network security, trustworthy computing and security for social networks. More specifically, he is interested in trust establishment, trustworthy code execution in the presence of malware, and how to design secure next-generation networks.

"Underwater Sensor Networks: Random Access and Compressive Sensing" by M. Stojanovic, Northeastern University

Wireless information transmission through the ocean is one of the enabling technologies for the development of future ocean-observation systems, whose applications range from oil industry to aquaculture and include gathering of oceanographic data, pollution control, climate recording, and transmission of images from remote sites. Underwater wireless communications are usually established using acoustic waves, since electro-magnetic waves propagate only over very short distances. However, acoustic communications are limited by three factors: low bandwidth, low speed of sound, and poor quality of the physical link. These constraints yield a difficult communication channel, while additional, system-level constraints come from the limited battery supply (re-charging is difficult underwater) and the half-duplex operation of existing acoustic modems. For networks that are deployed for long-term monitoring of environmental phenomena, it is crucial to design an efficient data gathering scheme that prolongs the system’s life-time. To this end, we exploit the sparse nature of the monitored field and design a random access / compressive sensing (RACS) scheme in which the sensors transmit at random to a fusion center which then reconstructs the field. We provide an analytical framework for system design that captures packet collisions due to random access, as well as packet errors due to communication noise. Through analysis and examples, we demonstrate that recovery of the field can be attained using only a fraction of the resources (energy per bit, bandwidth) used by a conventional TDMA network, while employing a scheme that is simple to implement, requires no synchronization or scheduling, and no downlink feedback.

Bio
Milica Stojanovic graduated from the University of Belgrade, Serbia, in 1988, and received the M.S. and Ph.D. degrees in electrical engineering from Northeastern University, Boston, MA, in 1991 and 1993. After a number of years with the Massachusetts Institute of Technology, where she was a Principal Scientist, she joined the faculty of Electrical and Computer Engineering Department at Northeastern University in 2008. She is also a Guest Investigator at the Woods Hole Oceanographic Institution, and a Visiting Scientist at MIT. Her research interests include digital communications theory, statistical signal processing and wireless networks, and their applications to underwater acoustic communication systems. Milica is a Fellow of the IEEE and serves as an Associate Editor for the IEEE Journal of Oceanic Engineering and the IEEE Transactions on Signal processing.

Link to Dr. Stojanovic's presentation slides.

http://eceweb1.rutgers.edu/lecture/slides/biology-schneider.pdf