2009-2010 CSCE 681 Abstracts
SPRING 2010 CSCE 681 Abstracts
CSCE 681 Graduate Seminar:
Sparse Spherical Arrangements and Geometric Algorithms for Molecular Structures
Dr. Dan Halperin
Professor
Tel Aviv University
4:10 p.m., Wednesday January 20, 2010
Room 124, Bright Building
Abstract
We developed algorithms for several problems in structural bioinformatics, among them algorithms for maintaining molecular surfaces under conformational changes, identifying channels in the complement of molecules, and simulating molecular motions. The efficiency of the algorithms relies in part on favorable geometric and kinematic properties of (common models of) molecules. We describe these properties and how the algorithms exploit them. More information can be found at http://acg.cs.tau.ac.il/projects.
Biography
Dan Halperin received his Ph.D. in Computer Science from Tel Aviv University in 1992. He then spent three years at the Computer Science Robotics Laboratory at Stanford University. In 1996 he joined the Department of Computer Science at Tel Aviv University, where he is currently a full professor and for two years until September 2006 was the department chair. Halperin's main field of research is Computational Geometry, with a focus on applied aspects and in particular on robustness and precision issues. Application areas he is currently interested in include robotics and automated manufacturing, algorithmic motion planning, and structural bioinformatics.
CSCE 681 Faculty Contact: Dr. Nancy Amato (amato [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Optimizing for Multiple Personalities on the Convey Hybrid-Core System
Robert Metzger
Senior Engineer
Convey Corporation
4:10 p.m., Monday January 25, 2010
Room 124, Bright Building
Abstract
The Convey HC-1 hybrid-core computer combines an Intel®64 processor with a coprocessor that supports instruction set architectures (multiple personalities) that can be created to accelerate individual algorithms. Each personality contains the same base set of scalar instructions. Personalities differ in terms of vector instructions, data types, algorithm-specific instructions, and hardware resources and speeds.
The Convey compiler suite is based on the Open64 compilers for C, C++, and Fortran 95. The optimizer has been enhanced to include a table-driven vectorizer, which applies transformations based on the availability of instruction set features of the target personality.
This presentation describes the Convey architecture, the structure of the vectorizer, the interface between the vectorizer and the personality file, the user-visible features provided for controlling the vectorization process, and source code examples of usage.
Biography
Robert Charles Metzger is currently a senior engineer who designs optimizing compilers at Convey Computer Corp. Convey is a venture-capital funded company in Richardson, TX, which provides adaptive, energy-efficient hybrid-core computers for the high performance computing marketplace.
During his career, Metzger has developed and managed the development of compilers, programming tools, and system administration software for high performance computers at Convex Computer Corp. and the Hewlett-Packard Company. He is the author of Debugging by Thinking: A Multidisciplinary Approach (Elsevier Digital Press, 2004) and co-author of Automatic Algorithm Recognition and Replacement: A New Approach to Program Optimization, (MIT Press, 2000).
CSCE 681 Faculty Contact: Dr. Lawrence Rauchwerger (rwerger [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Concurrency Programming in Visual Studio 2010
Terry Leeper
Principal Architect, WinC++
Microsoft
4:10 p.m., Wednesday, January 27, 2010
Room 124, Bright Building
Abstract
Prior to Visual Studio 2010, correctly introducing concurrency into libraries and applications has been difficult, time consuming, and error-prone. However, as the hardware industry shifts towards multi-core and manycore processors, the key to high-performance applications is parallelism. Visual Studio 2010, Visual C++ 10, and the .NET Framework 4 offer solutions to help make writing parallelized applications significantly easier. In this session, we explore new runtimes, libraries, and tools available in Visual Studio 2010, providing an in-depth look at the next generation of parallel programming on the Microsoft platform.
Biography
tba
CSCE 681 Faculty Contact: Dr. Lawrence Rauchwerger (rwerger [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Partial Automation of the Cartographic Workflow for Disaster Situations
Richard Smith
Visiting Assistant Professor
Texas A&M University, Corpus Christi
4:10 p.m., Monday, February 1, 2010
Room 124, Bright Building
Abstract
Geographic Information Systems (GIS) have shown to be extremely useful in disaster planning and recovery situations. A GIS provides decision makers with the capability of real-time collection, analysis, visualization and dissemination of spatial and non-spatial information through mapping products. Successful cartographic GIS products are based on their ability to convey mapping information accurately and clearly. Designing a successful cartographic product takes time as it is often an iterative process of choosing colors, line weights, symbols, patterns, and arranging data layers. However, in a disaster situation, time is critical, and, thus, cartographic decisions are often relegated to "when there is time!" This sidelining of cartography and the enormous demand for well-designed maps are contradictory. The flow of information holds a higher priority when a disaster hits and cartography obtains a lower priority. This research examines a potential solution to this problem utilizing experiences and lessons learned during the recovery and mapping of the NASA Columbia Space Shuttle debris. An attempt at reducing time requirements for cartographic decisions is investigated. This research is accomplished through data mining of metadata, creation and utilization of existing cartographic ontologies, allowing for re-use of existing cartographic decisions in similar disaster situations and leveraging the inferential capabilities of artificial intelligence and expert systems.
Biography
Rick Smith is a Visiting Assistant Professor of Geographic Information Science in the Department of Computing Sciences at Texas A&M University Corpus Christi. He is a PhD candidate in Geography at the University of Georgia, holds a Master of Science in Computer Science and a Bachelor of Science in Geographic Information Science from Texas A&M University Corpus Christi. Rick's research interests are in systems integration, and bringing GIS to the masses through custom application development and the democratization of cartography. He has designed and implemented custom GIS software and authored maps for both the public and private sector. Such projects include designing and implementing a Gang Tracking Database with mapping functionality for the Nueces County Sheriff's Department, authoring a web-based mapping application for NASA while working with Lockheed Martin Space Operations, mapping the Columbia Shuttle debris for NASA's Mishap Investigation Team, developing embedded computer systems for the logging of near real-time weather and tide information for the Texas Coastal Ocean Observation Network, developing a mobile GIS solution for rapid disaster response for NASA, authoring handheld software streamlining geodetic leveling for the National Geodetic Survey, authoring embedded software for automatic logging and reporting of weather station information, and designing and implementing an autonomous terrain mapping system for industrial applications. Rick has co-authored multiple peer-reviewed articles, presented at numerous national and international conferences, and published several maps in various mediums including the cover on the most recent edition of the Journal of Geography. Rick's current research centers on the democratization of cartography through the partial automation of the cartographic workflow through data mining metadata, developing a cartographic ontology, and computer-based inference techniques.
CSCE 681 Faculty Contact: Dr. Robin Murphy (murphy [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Evolving a language in and for the real world
Dr. Bjarne Stroustrup
College of Engineering Chair in Computer Science Professor
The Department of Computer Science and Engineering
Texas A&M University
4:10 p.m., Wednesday, Feb 10, 2010
Room 124, Bright Building
Abstract
This is a talk about the design and evolution of C++. As all programming languages, C++ owes a lot to earlier languages and evolves through the blending of ideas. This talk tries to answer some common questions about that success of C++: why did it succeed? At what did it succeed? How did it maintain a steady course over more than 25 years? At what did it not succeed? The scale of C++'s success was unanticipated and its continuing strength has left many language and business theorists puzzled, so explanations are required. Given the long time span involved and because no large system can be designed with 20-20 hindsight, a historical perspective is an essential part of any answer. A combination of technical, philosophical, and sociological issues must be considered. This talk focuses on the design aims of C++ and my philosophy of incremental language evolution relying on feedback loops. It goes into some detail on the ISO standards process which has been central to the stability and evolution of C++. It gives a few – but only very few – illustrative code examples.
Biography
Bjarne Stroustrup is the designer and original implementer of C++ and the author of "The C++ Programming Language", "The Design and Evolution of C++, "Programming — Principles and Practice using C++", and many other publications. Dr. Stroustrup is the College of Engineering Chair Professor in Computer Science at Texas A&M University. He's a member of the National Academy of Engineering, an AT&T Fellow, an IEEE fellow, and an ACM fellow. His research interests include distributed systems, design, programming techniques, software development tools, and programming languages. He is actively involved in the ANSI/ISO standardization of C++.
CSCE 681 Faculty Contact: Dr. Nancy Amato (amato [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Clifford Subsystem Codes
Dr. Andreas Klappenecker
Associate Professor
Department of Computer Science and Engineering
Texas A&M University
4:10 p.m., Wednesday, March 3, 2010
Room 124, Bright Building
Abstract
Quantum computing is one of the most exciting developments in computer science. It is conjectured that quantum computers will be able to solve certain problems faster than any classical computer. It is quite encouraging that some small scale quantum computers exist, but further progress is needed to enable fully-scalable fault-tolerant quantum computing.
In fault-tolerant quantum computing, the quantum information is encoded with the help of a quantum error-correcting code. All operations are performed on the encoded quantum bits. Even though the error-correction itself is faulty, one can achieve reliable computation with a modest overhead as long as the error rate of the quantum operations is below a threshold.
The speaker has developed with others the theory of subsystem codes. They have the advantage that they underlie fewer constraints than other codes that have been used so far in fault-tolerant quantum computation. Subsystem codes can offer encoding algorithms that allow for noisy auxiliary quantum bits, can offer better decoding algorithms and sometimes allow one to realize fault-tolerant operations more efficiently than with other methods. The talk will outline the evolution of Clifford subsystem codes, from inception and arrested development to the most versatile class of quantum codes in existence today. The talk will outline how quantum systems of mixed dimensions can be protected by Clifford subsystem codes.
Biography
Andreas Klappenecker is Associate Professor of Computer Science at Texas A&M University. In 1998, he received his Ph.D. in Computer Science from the University of Karlsruhe, Germany. He joined Texas A&M University in 1999 as a Visiting Assistant Professor of Mathematics, and worked shortly as a Research Associate at the University of Karlsruhe before accepting his current position. Dr. Klappenecker received a National Science Foundation CAREER award and a TEES Select Young Faculty award in 2004, was named Halliburton Faculty Fellow in 2007, Fellow-at-Large of the Santa Fe Institute in 2000, and received highest distinction for his Ph.D. thesis.
CSCE 681 Faculty Contact: Dr. Nancy Amato (amato [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Inferring Private Information Using Social Network Data
Dr. Murat Kantarcioglu
Assistant Professor
The University of Texas at Dallas
4:10 p.m., Wednesday, March 10, 2010
Room 124, Bright Building
Abstract
On-line social networks, such as Facebook, are increasingly utilized by many users. These networks allow people to publish details about themselves and connect to their friends. Some of the information revealed inside these networks is private and it is possible that corporations could use learning algorithms on the released data to predict undisclosed private information. In this talk, we discuss how to launch inference attacks using released social networking data to predict undisclosed private information about individuals. We then explore the effectiveness of possible sanitization techniques that can be used to combat such inference attacks under different scenarios. In addition, we discuss fine grained access control mechanisms that leverage semantic web technologies for protecting sensitive social network data.
Biography
Dr. Murat Kantarcioglu is currently an assistant professor of computer science at University of Texas at Dallas. He had a Ph.D. degree from Purdue University in 2005. He received his master's in Computer Science from Purdue University in 2002 and his bachelor degree in computer engineering from METU, Ankara, Turkey in 2000. He was also a recipient of NSF CAREER Award.
Dr. Kantarcioglu's research focuses on creating technologies that can efficiently extract useful information from any data without sacrificing privacy or security. Recently, he has been working on security and privacy issues raised by data mining, privacy issues in social networks, privacy issues in health care, risk and incentive issues in assured information sharing, and use of data mining for fraud detection and homeland security. His research has been funded by grants from NSF, NSA, AFOSR, NIH, ONR and IARPA.
CSCE 681 Faculty Contact: Dr. Guofei Gu (guofei [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
High Performance Cloud Computing: An Emerging Strategy for Scientific Computing
Dr. Ekpe Okorafor
African University of Science & Technology &
Visiting Professor
The University of Texas at Dallas
4:10 p.m., Monday, March 22, 2010
Room 124, Bright Building
Abstract
Scientific computing often requires the availability of a massive number of computers for performing large scale experiments. Traditionally, high-performance computing solutions and installed facilities such as clusters and super computers have been employed to address these needs. Cloud computing provides scientists with a completely new model of utilizing the computing infrastructure.
The infrastructure services (Infrastructure-as-a-service), provided by these cloud vendors, allow any user to provision a large number of compute instances. However, scientific computing is typically characterized by complex communication patterns and requires optimized runtimes. This talk will outline the challenges in exploiting cloud computing technologies for scientific computing, and its potential applications, especially, in the context of the peculiar infrastructure environment in developing countries characterized by low bandwidth, low computing capacity and unreliable service.
Biography
Dr. Ekpe Okorafor received the M.Sc. and Ph.D. degrees in Electrical and Computer Engineering from Texas A&M University in 2002 and 2005, respectively. He is currently, a faculty at the African University of Science & Technology (AUST), Abuja. He has worked in leading research labs including IBM Watson Research and Almaden Research Centers and has managed a large Information Technology department.
Dr. Ekpe Okorafor's research areas include network traffic modeling, high-performance computing, large complex computing systems, virtualization, grid/cloud computing, performance modeling and optimization. He has authored and co-authored many conference papers and journal publications.
CSCE 681 Faculty Contact: Dr. Nancy Amato (amato [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Venture Investing 2010
Jeanne Bayless
Managing Partner
Star Ventures USA
4:10 p.m., Wednesday, April 14, 2010
Room 124, Bright Building
Abstract
The presentation will provide a reflection into the history of venture capital and why it continues to be an important driver for the US economy. It will also provide an examination of the structure of venture firms, how they operate from investing to fundraising, and how venture capitalists assess a potential investment. The speaker will also discuss one of her current investments, Orametrix, Inc., that has used computer technology to influence the field of orthodontics.
Biography
Ms. Bayless joined STAR Ventures in 2002 as Managing Partner of the U.S. office. Before joining STAR, she served as Director of Strategic Investments for Intel Capital. Prior to her tenure at Intel Capital, Ms. Bayless was the Founder, CEO and President of AnswerSoft, a startup that developed software for the automation of inbound call center operations. AnswerSoft was acquired by Davox Corporation in 1998, and Ms. Bayless assumed the role of VP of Marketing after the acquisition. After Davox she assumed the role of VP of Marketing at NetBoost Corporation, a startup that developed solutions for the acceleration of network applications. NetBoost was acquired by Intel Corporation in 1999. Ms. Bayless is currently an active Board member of the following companies: CedarPoint Communications, OraMetrix, Sipera Systems, and Teneros. Former Board positions include Acopia Networks (acquired by F5), Broadbus (acquired by Motorola) and GrooveMobile (acquired by LiveWire). She also holds Board observer roles at Mintera Corporation, Medical Present Value, Everyday Health, and Xtera Communications. She serves on the Board of the Metroplex Technology Business Council, Board of the North Texas Regional Center of Innovation and Commercialization and the Advisory Boards of the Entrepreneurs' Foundation of North Texas and the University of Texas at Dallas School of Management. Ms. Bayless holds an M.S. in Management from The University of Texas at Dallas and a B.S. in Computer Science from Texas A&M University.
CSCE 681 Faculty Contact: Dr. Nancy Amato (amato [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Endpoint Prediction Using Motion Kinematics
Dr. Edward Lank
Assistant Professor
The University of Waterloo
4:10 p.m., Monday, April 19, 2010
Room 124, Bright Building
Abstract
Point-and-click interaction (pointing) is frequent task in modern graphical user interfaces (GUIs), so even a marginal improvement in pointing performance can have a dramatic impact on user efficiency. Many researchers have studied pointing facilitation techniques, but these techniques typically make one of two assumptions. Either potential targets are relatively sparse on the display or there exists some technique for predicting a user's target, or endpoint, during their movement toward a target.
In this talk, I will describe our work on endpoint prediction using speed signatures of human motion. This work combines two laws of human motion from psychology research, the stochastic optimized sub-movement model and the minimum jerk law, and develops an approach that makes use of these laws to extrapolate motion endpoint in computer interfaces during a gesture. Using this technique, we show that it is possible to predict a region of interest on the computer display, to expand groups of targets within that region, and to combine our technique with other prediction strategies such as command frequencies or user task modeling. Together, these results present an important step forward in realistic pointing facilitation techniques in modern GUIs.
Biography
Dr. Edward Lank is an Assistant Professor in the David R. Cheriton School of Computer Science at the University of Waterloo. His research is in the area of Human-Computer Interaction (HCI), including applications of tablet computing, the study of motion kinematics in interfaces, and the design of pervasive computing applications. Prior to joining the faculty at Waterloo, Dr. Lank was an Assistant Professor of Computer Science at San Francisco State University, was a research intern at the Palo Alto Research Center in the Perceptual Document Analysis Area; was Chief Technical Officer of MediaShell Corporation, a Queen's University research start-up; and was an Adjunct Professor in the Department of Computing and Information Science at Queen's University. He received his Ph.D. in Computer Science from Queen's University in Kingston, Ontario, Canada in 2001 under the supervision of Dr. Dorothea Blostein. He also holds a Bachelor's Degree in Physics with a Minor in Computer Science from the University of Prince Edward Island.
CSCE 681 Faculty Contact: Dr. Tracy Hammond (hammond [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Manifold Learning in Human-Robot Teams
Dr. Chad Jenkins
Assistant Professor of Computer Science
Brown University
4:10 p.m., Wednesday, April 21, 2010
Room 124, Bright Building
Abstract
A principal goal of robotics is to realize embodied systems that are effective collaborators in human endeavors pursued in the physical world. Human-robot collaborations can occur in a variety of forms, including autonomous robotic assistants, mixed-initiative robot explorers, and augmentations of the human body. For these collaborations to be effective, human users must have the ability to realize their intended behavior into actual robot control policies. At run-time, robots should be able to manipulate an environment and engage in two-way communication in a manner suitable to their human users. Further, the tools for programming, communicating with, and manipulating using robots should be accessible to the diverse sets of technical abilities present in society. Towards the goal of effective human-robot collaboration, learning from demonstration (LfD) has emerged as a central theme of our work for the natural instruction of autonomous robots by human users. In robot LfD, desired cognitive functions for a robot (perception, decision making, or motion control) are implicit in human demonstration rather than explicitly coded in a computer program.
In this talk, I will present our work into learning priors from human demonstration for robot perception and control using manifold-based dimension reduction. My specific focus will be the development and application of manifold learning algorithms to estimate subspace priors for spatial and time-series data generated by humans. I will discuss our approach to spatio-temporal dimension reduction in the context of manifold learning. Using manifold learning, results will be presented from learning priors for: 1) classifying tactile signatures to recognize successful grasps on the NASA Robonaut, 2) providing low-dimensional control spaces for neural prosthetics, 3) learning motion primitives from human movement data, and 4) extracting kinematic models and poses from multi-view video of human performance. Our approach to learning priors will be cast in the broader context of policy learning and computational models for communication in multi-robot multi-human systems.
Biography
Odest Chadwicke Jenkins, Ph.D., is an Assistant Professor of Computer Science at Brown University. Prof. Jenkins earned his B.S. in Computer Science and Mathematics at Alma College (1996), M.S. in Computer Science at Georgia Tech (1998), and Ph.D. in Computer Science at the University of Southern California (2003). Prof. Jenkins was selected as a Sloan Fellow and a Kavli Fellow in 2009. He is a recipient of the Presidential Early Career Award for Scientists and Engineers (PECASE) for his work in physics-based human tracking. He has also received Young Investigator awards from the Office of Naval Research (ONR) for his research in learning dynamical primitives from human motion, the Air Force Office of Scientific Research (AFOSR) for his work in manifold learning and multi-robot coordination and the National Science Foundation (NSF) for robot learning from multivalued human demonstrations. His research addresses problems in robot learning and human-robot interaction, primarily focused on robot learning from demonstration, as well as topics in computer vision, machine learning, and computer animation.
CSCE 681 Faculty Contact: Dr. Dylan Shell (dshell [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Foreign accent conversion through voice morphing
Dr. Ricardo Gutierrez-Osuna
Associate Professor
Department of Computer Science and Engineering
Texas A&M University
4:10 p.m., Monday, April 26, 2010
Room 124, Bright Building
Abstract
Despite years or decades of immersion in a new culture, older second-language learners typically speak with a so-called "foreign accent," sometimes despite concerted efforts at improving pronunciation. A number of studies have suggested that it would be beneficial for such learners to be able to listen to their own voices producing native-accented speech. As a step towards this goal, we are developing speech processing methods to modify the perceived accent of utterances from foreign speakers of English. Our approach consists of decomposing the speech signal of a learner and a teacher into two components: one that carries the speakers' voice quality and a second one that contains their linguistic gestures. By combining the voice quality of the learner with the linguistic gestures of the teacher, we can then generate a "morphed" voice that is perceived to be like that of the foreign speaker but has the accent of the native speaker.
In this talk, I will review previous work on computer assisted pronunciation training, describe our speech processing method for foreign accent conversion, and present results from perceptual evaluations of our accent morphs. I will also describe additional work we are conducting in the areas of active sensing, wearable sensors and face perception.
Biography
Ricardo Gutierrez-Osuna received a B.S. degree in Electrical Engineering from the Polytechnic University of Madrid (Spain) in 1992, and M.S. and Ph.D. degrees in Computer Engineering from North Carolina State University in 1995 and 1998, respectively. He is currently an Associate Professor in the Department of Computer Science and Engineering at Texas A&M University. He has broad research interests in intelligent sensors, machine learning, neuromorphic computation, and models of human perception.
CSCE 681 Faculty Contact: Dr. Nancy Amato (amato [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Generic Programming for Parallelism: From Concepts to Tools
Dr. Joel Falcou
Assistant Professor
Department of Computer Science
University Paris XI, France
4:10 p.m., Tuesday, May 4, 2010
Room 124, Bright Building
Abstract
As emerging architectures becomes more powerful, they also become more complex to handle for everyday's developers.
In this context, building tools for a large variety of upcoming architectures (GPGPU, many-cores, SMP clusters) is a major road-block on mastering these architectures. This presentation looks at how generic and generative programming idioms allow for both expressive and efficient implementation of such parallel programming tools.
Biography
Joel Falcou is a Computer Science assistant professor at University Paris XI, France. His work investigates how modern development idioms like generic or generative programming can be applied to solve modern parallel programming challenges and how such idioms can be formalized.
CSCE 681 Faculty Contact: Dr. Lawrence Rauchwerger (rwerger [at] cse.tamu.edu)
FALL 2009 CSCE 681 Abstracts
CSCE 681 Graduate Seminar (only required for new graduate students):
Graduate Orientation I: Overview of Department Resources & Contacts, Honor Code, and Student Organizations
MANDATORY FOR NEW GRAD STUDENTS (but not other CSCE 681 students)
4:10-6:00 p.m., Monday August 31, 2009
Room 124, Bright Building
Abstract
This meeting will concentrate on the essentials the students will need to settle in. It will include an introduction to departmental administration (staff who's who, payroll, mailboxes, phones), the computing resources (computer use/accounts, printer quotas, lab access/tours), the academic advising staff and resources, the TAMU honor system, TAMU Libraries, Graduate Teaching Academy, Student Engineers' Council and relevant student organizations (CSGSA,
AWICS, TACS (TAMU ACM and IEEE student chapter), UPE and TAGD).
MANDATORY FOR NEW GRAD STUDENTS (but not other CSCE 681 students)
CSCE 681 Graduate Seminar (required for all new graduate students and all CSCE 681 students):
Graduate Orientation II: Current Grad & Faculty Panels, Poster Session, & PIZZA!
MANDATORY FOR NEW GRAD STUDENTS. Counts as CSCE 681 seminar for CSCE 681 students.
4:10-6:00 p.m., Wednesday September 2, 2009
Room 124, Bright Building
Abstract
- 4:10-4:40 p.m. - Student Panel: Current grad students share tips about how to succeed in graduate school.
- 4:40-5:10 p.m. - Faculty Panel: Faculty share their ideas about what they are looking for in a graduate student.
- 5:10-6:00 p.m. - Pizza & Current Student Poster Session - new students can meet current grads and learn about ongoing research projects.
MANDATORY FOR NEW GRAD STUDENTS and CSCE 681 STUDENTS
CSCE 681 Graduate Seminar:
Macroscopic Modeling for Robot Swarms
Dr. Dylan Shell
Assistant Professor
Department of Computer Science and Engineering
Texas A&M University
4:10 p.m., Monday September 7, 2009
Room 124, Bright Building
Abstract
There are many potential future applications of multi-robot systems: teams of fire-fighting robots to limit casualties, groups of robots to clear land mines, and even nano-robot swarms designed to destroy toxins from within. But to be effective robots must coordinate their efforts in order to act as a synergistic whole. This talk will outline some of the broad problems in multi-robot systems, what particular challenges make them distinct from other multi-agent systems, and ultimately why they're interesting and worthy of scientific study. In doing so, we'll briefly look at how far this particular research area has come.
In this talk I will also describe a particular class of swarm robot systems with hundreds of robots and will show that the large size of such swarms makes them amenable to statistical analysis. I will describe a framework based on equilibrium thermodynamic and statistical mechanical methods that enables a controller synthesis methodology for homogeneous robot swarms. This allows system designers think about controller synthesis as the problem of combining formally defined macroscopic templates rather than as manipulation of low-level controllers. In addition to being a step toward a principled design methodology for multi-robot systems, an important contribution of the work is to show that ergodic processes are sufficiently powerful to produce useful coordinated behavior in groups of robots.
Biography
Dylan Shell is an assistant professor computer science and engineering at Texas A&M University in College Station, Texas. He received his BSc degree in computational & applied mathematics and computer science from the University of the Witwatersrand, South Africa, and his M.S. and Ph.D in Computer Science from the University of Southern California. He took a position as Postdoctoral Research Associate in the USC Interaction lab in 2008, before joining Texas A&M. His research aims to synthesize and analyze complex, intelligent behavior in distributed systems that exploit their physical embedding to interact with the physical world. He studies both natural (social insect and human crowd) and synthetic systems (sensor networks and multi-robot swarms) by using techniques that model behavior across multiple scales. Currently his approach is to treat broad (even qualitative) models of group behavior as descriptions of the system's macroscopic information processing capabilities. Although his focus is on multi-robot applications of these ideas. His broader research interests include unconventional and statistical models of computation, cooperation and competition in artificial systems, emergence, statistical mechanics and thermodynamics of finite systems.
CSCE 681 Faculty Contact: Dr. Nancy Amato (amato [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Searching String Collections for the Most Relevant Documents
Dr. Jeffrey S. Vitter
Professor
The Department of Computer Science and Engineering
Texas A&M University
4:10 p.m., Monday September 14, 2009
Room 124, Bright Building
Abstract
This talk addresses how to search huge document collections and, for a given query pattern, find the most relevant documents that contain that pattern. "Most relevant" may mean the documents that contain the greatest number of instances of the pattern or the documents with the highest Page Rank (such as used by Google). Inverted indexes do not handle general pattern search, and suffix trees and suffix arrays are too expensive in terms of space usage, and in addition, they require the finding of every occurrence of the pattern, which can be very expensive when the number of pattern occurrences is much larger than the number of documents. We improve upon the results of Muthukrishnan. In addition, we show how to reduce the data structure size so that it is proportional to the size of the document collection.
Joint work with Wing-Kai Hon and Rahul Shah
Biography
Jeff Vitter is professor of computer science and engineering at Texas A&M University in College Station, Texas. From 2008 to 2009, he served as is Provost and Executive Vice President for Academics at Texas A&M University. Before coming to Texas A&M in August 2008, he served six years as the Frederick L. Hovde Dean of the College of Science and Professor of Computer Science at Purdue University in West Lafayette, Indiana. After earning a bachelor's degree in mathematics with highest honors from the University of Notre Dame in 1977 and a PhD in computer science from Stanford University in 1980, he began his academic career as assistant professor of computer science at Brown University in 1980 where he progressed through the faculty ranks and served in various leadership roles. From 1993 to 2002, Dr. Vitter was the Gilbert, Louis, and Edward Lehrman Professor of Computer Science in the College of Arts and Sciences at Duke University in Durham, North Carolina. He served as chair of the Department of Computer Science at Duke from 1993-2001 and as co-director and a founding member of Duke's Center for Geometric and Biological Computing from 1997-2002. While serving on the faculty at Duke, he earned an M.B.A. from the Fuqua School of Business at Duke in 2002.
Dr. Vitter has been named a Guggenheim Foundation Fellow, a Fellow of the Association for Computing Machinery, a Fellow of the Institute of Electrical and Electronics Engineers, a National Science Foundation Presidential Young Investigator, a Fulbright Scholar, and an IBM Faculty Development Awardee. He has over 250 book, journal, and conference publications; his Google Scholar h-index is 54. He authored the book Algorithms and Data Structures for External Memory (now Publishers, 2008), which covers the I/O field he helped found. He has also coauthored the books Efficient Algorithms for MPEG Video Compression (Wiley & Sons, 2002) and Design and Analysis of Coalesced Hashing (Oxford University Press, 1987). He is coeditor of the collections External Memory Algorithms and Algorithm Engineering and co-holder of patents in the areas of external sorting, parallel I/O, prediction, and approximate data structures. Editorial board memberships have included Algorithmica, Communications of the ACM, IEEE Transactions on Computers, Theory of Computing Systems (formerly Mathematical Systems Theory: An International Journal on Mathematical Computing Theory), and SIAM Journal on Computing; in addition, he has edited several special issues.
One theme in Dr. Vitter's research and teaching is how to alleviate the I/O bottleneck between fast internal memory and slow external storage (such as disk) that can occur when processing massive data sets. He is credited as a founder of the field of external memory algorithms. He has pioneered the development of efficient external memory algorithms in several domains, including geographic information systems (GIS) and spatial databases, sorting, text and string indexing, matrix computations, graph traversal, range search, data mining, and a variety of computational geometry and combinatorial problems. A related interest is how to take advantage of parallel disks or parallel hierarchical memories, in which communication with each parallel memory device can occur simultaneously. He is involved in algorithm engineering using the TPIE system (Transparent Parallel I/O programming Environment). Another aspect of Dr. Vitter's work involves novel machine learning and prediction mechanisms based upon data compression and locality, using the principle that the more compressible a sequence is, the more predictable it is. Examples include algorithms for caching, prefetching, data streams, database query optimization, data mining, and resource management in mobile computers. He introduced the important use of wavelets in database applications and through that work was co-recipient of the 2009 SIGMOD Test of Time award. He has worked on efficient approaches to image, video, and text compression. He currently works on compressed data structures for searching, where the goal is to use a small amount of space equal to the entropy of the input data, yet still achieve fast search time. Previously, fast data structures for text indexing (such as suffix trees and suffix arrays) required several times more space than the data being indexed! Other interests include randomized, parallel, and incremental algorithms for computational geometry, graphics, random sampling, and random variate generation.
CSCE 681 Faculty Contact: Dr. Nancy Amato (amato [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
The 15 Year Cycle of Robotics: Commercializing Robots
Dr. Allan Branch
Founder and Chief Executive Officer
Denning Branch International
4:10 p.m., Monday September 21, 2009
Room 124, Bright Building
Abstract
After perhaps a half century of global robot development in academic and commercial sectors, with substantial progress in many key technologies such as Artificial Intelligence and microprocessors, significant media exposure and exposure through popular culture such as toys and movies, large rounds of funding from VC's and agencies like DARPA, and at times premium clients like the military, it is fair to say that robot markets have nevertheless not lived up to their oft touted promises. This paper examines why and presents key factors holding the industry back. The work is based on two decades of development, manufacturing and marketing activities from within a private commercial enterprise, with technical research sponsored in part by the Australian Federal and Tasmanian State Governments, but a company nevertheless reliant essentially on its survival from sales of working robots that customers wanted to buy. One interesting result is the identification of a repetitive phenomenon called, "The 15 Year Cycle of Robotics" illustrating the repetition of mistakes. More importantly, there is another conclusion; it means that there is something else other than the inventiveness and genius behind technology development that determines its efficacy and influence. Any successful entrepreneur knows this. In the real world, finance, personalities, market demands, networks, chance, luck, and such, really do play a big role in the way our society progresses, or fails to progress. These observations are well known and documented. In the case of mobile robotics though, it is not clear that they are the reasons for the persistent commercial failure. Opportunity and resources have never been lacking. So is there something unique about robotics marketing, and therefore problems and solutions that have escaped those engineers and scientists who have ventured into the business world? The answer is a definite yes.
Biography
Allan Branch is Australian and lives between New York City and Munich, Germany. He is CEO and Chairman of Denning Branch International. He attended the School of Medicine at the University of Tasmania before entering into what became an 18 year career inventing and manufacturing smart mobile robots starting with the release of the Tasman Turtle educational Logo robot in 1979 and culminating with the pioneering robotics corporation Denning Mobile Robotics, Inc. which he turned around from commercial failure through the years 1993 to 1997. Since then he has had a second career driving international corporate turnarounds and is currently completing his first book "Save the Day, How to Fix any Failed or Failing Company without Downsizing." During his robotics years, he frequently "escaped" to interesting secondments including a period with Commodore Business Machines in Texas (1983/84), head of General Electric Plastics' robotics Florbot project in Pittsfield, Massachusetts (1988/89), a year with Silicon Valley legend Nolan Bushnell at Axlon (1987) and head of the D'Entrecasteaux domestic floorcare robot project for Moulinex in Caen, France (1991-93). He was the Invited Visiting Scientist to Hans Moravec's laboratory at CMU in 1989/90 where his PhD and MBA career equivalencies were recognized for his development of the autonomous navigation system Parametric Mapping and the financial analysis tool Historical Business Process Trend Analysis, both proprietary technologies that have facilitated his current work consulting to academia and corporations in healthcare, biotech, computer science, agriculture and FMCG among others. Allan's early career included interesting positions in teams pioneering organ transplants, artificial hearts, productivity enhancement for the Australian Federal Government and designing juke boxes. He has held top executive appointments in technology companies listed on NASDAQ, London Stock Exchange, French CAC40 and Australian ASX, developed several faculty business plans for Sydney University and been CEO and Chairman of the Board of RNAPS, a 170 year old Royal Society. His professional interests include Corporate Governance, Technology Commercialization and playing with clever children's toys.
CSCE 681 Faculty Contact: Dr. Robin Murphy (murphy [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Control of Unmanned Aircraft Systems for Communication and Atmospheric Sensing Applications
Dr. Eric Frew
Assistant Professor
Research and Engineering Center for Unmanned Vehicles (RECUV)
University of Colorado at Boulder
4:10 p.m., Monday September 28, 2009
Room 124, Bright Building
Abstract
Wireless networked communication plays a key role in the coordination and control of robotic sensor networks. Most work in the field of cooperative control assumes simple graph-theoretic models to describe connectivity among neighboring agents. Unfortunately this model is overly simplistic and does not capture wireless network behaviors in realistic communication environments. In particular, wireless communication is not easily described by sets of "links", but rather, communication capability changes as a function of separation distance between agents, and fading (noise) effects are difficult to predict. This presentation describes a framework for controlled mobility of multi-vehicle networks performing active sensing tasks in realistic communication environments.
An information-theoretical formulation of active sensing tasks will be presented that combines sensing and communication. Optimal robot motion is derived using this new formulation. Mitigation of noise and other unmodeled effects is achieved through decentralized model-free adaptive control techniques to improve information flow relative to the positions determined by the initial optimization. Experimental results obtained from the Heterogeneous Unmanned Aircraft System (HUAS) at the University of Colorado at Boulder are presented. Compared to other UAS test-beds operated by university and government research institutions, a key innovation of HUAS is the presence of a multi-tier airborne mobile ad-hoc network.
Time permitting; this presentation will also describe current efforts developing planning algorithms for motion in strong wind and current fields. Small unmanned aerial and underwater vehicles are particularly susceptible to strong wind/currents where the air/water speed is greater than the vehicle speed. New planning techniques that can be run in real-time are needed to account for these limitations. The motivating example of unmanned aircraft deployment in severe convective storms will be discussed.
Biography
Dr. Eric W. Frew is an assistant professor in the Aerospace Engineering Sciences Department at the University of Colorado at Boulder where he is also a member of the Research and Engineering Center for Unmanned Vehicles (RECUV). He received his B.S. in mechanical engineering from Cornell University in 1995 and his M.S and Ph.D. in aeronautics and astronautics from Stanford University in 1996 and 2003, respectively. Prior to joining the CU Boulder faculty, he was a postdoctoral researcher at the UC Berkeley Center for Collaborative Control of Unmanned Vehicles where he oversaw the development and flight demonstrations of a fleet of three intelligent aerial platforms. Dr. Frew's research efforts focus on self-directed collaborative navigation of unmanned aircraft, exploitation of controlled mobility for integrating communication into multi-objective control, and optimal distributed sensing by teams of autonomous vehicles. He has over 60 journal, book chapter, and conference papers. In 2006 he was named Young Engineer of the Year by the AIAA Rocky Mountain Section in recognition of his "contributions to the field of unmanned aircraft systems in the areas of sensing, planning, and control." He received the NSF Faculty Early Career Development (CAREER) Award in 2009. A more complete bio can be found at http://recuv.colorado.edu/~frew.
CSCE 681 Faculty Contact: Dr. Dezhen Song (dzsong [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Wavefront Power Shifting via Thrifty Interconnection Network
Dr. Jian Li
Novel Systems Architecture Group
IBM Austin Research Laboratory
4:10 p.m., Monday October 5, 2009
Room 124, Bright Building
Abstract
Energy has become a key constraint for the design of modern data centers and large-scale systems. This paper presents two complementary techniques to manage the power consumption of large-scale systems with a packet-switched interconnection network. First, we propose Thrifty Interconnection Network (TIN), where the network links are activated and de-activated dynamically with little or no overhead by using inherent system events to trigger link activation or de-activation timely. Second, we propose Wavefront Power Shifting (WPS), a software-hardware hybrid approach, that dynamically shifts the power budget between the compute nodes and the interconnection network. WPS and TIN together activate and train the links in the interconnection network, just-in-time before the network communication is about to happen, and put them into a low-power mode thriftily when communication is finished, hence reducing unnecessary network power consumption. At the mean time, the compute nodes can absorb the extra power budget shifted from the thrifty network and increase their operating frequency for higher computing speed. Our simulation results on a set of real world HPC workload traces show that TIN and WPS together achieve on average 35% network power reduction, 6% system energy reduction, and 3% system performance improvement.
Given time, I will also introduce a few on-going projects at ARL system department, e.g., PERCS, Roadrunner, next generation HPC systems, data center power management, simulation technologies, etc, as well as some other papers/submissions I have co-authored recently, e.g., hybrid cache (ISCA'09, DATE'09), cache morhing, wireless NoC, stocastic confidence estimator, etc.
Biography
Jian Li is a research staff member at the Novel Systems Architecture group at IBM Austin Research Laboratory. He holds a Ph.D. degree in Electrical and Computer Engineering from Cornell University. He has worked in the areas of architectural support for power- and variation-aware computing, interconnection network design for high-performance computing systems, workload-driven three-dimensional (3D) integration architecture, architectural applications of non-volatile memory (NVM) and storage class memory (SCM), and energy-efficient interconnection networks, with a strong emphasis on entertaining his family members in Austin, Texas.
CSCE 681 Faculty Contact: Dr. Lawrence Rauchwerger (rwerger [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Exceptional Service in the National Interest
Dr. Guylaine M. Pollock
Senior Member, Technical Staff
Sandia National Laboratories
4:10 p.m., Wednesday October 7, 2009
Room 124, Bright Building
Slideshow: Sandia National Laboratories Labs Opportunities 2010
Slideshow: Sandia National Laboratories Labs Cyber Opportunities 2010
Abstract
Sandia National Laboratories, a government owned, contractor operated federally funded research and development center, began as the Z Division of Los Alamos in 1945, and became Sandia Laboratory in 1948 establishing itself as Sandia Corporation in 1949. Sandia's overarching purpose of providing "Exceptional Service in the National Interest" over the years has positioned Sandia as a multiprogram lab engaging in a broad spectrum of national security issues. This presentation provides a brief historical perspective of Sandia's service to the U.S. and presents SNL's core capabilities, responsibilities, mission areas, and primary technical staff disciplines before highlighting current technical program areas of importance to SNL. This overview includes a review of current interactions with Texas A&M University, and other opportunities and avenues for collaborative research and interactions with SNL, such as Fellowship Programs, Student Internships, etc.
Biography
Dr. Guylaine M. Pollock is a senior member of technical staff at Sandia National Laboratories. Previously, she was a research associate and instructor at Texas A&M University, receiving an IBM Federal Systems Division grant. She received a BS with academic distinction and highest honors in computer science/mathematics from East Texas State University, and was a student of computing science and operations research in Industrial Engineering at Texas A&M University, receiving a Ph.D. for her work there. She served as the 2000 President of the IEEE Computer Society and as the Division V IEEE Director for 2001-2003. She currently serves on the Engineering Advisory Council for Texas A&M University's Dwight Look College of Engineering. Guylaine has received numerous awards, including Notable Women of Texas, Upsilon Pi Epsilon membership, the society's Richard E. Merwin Scholarship, and a Gulf Oil Foundation Fellowship. Dr. Pollock received a personal letter of thanks from President Bill Clinton for work regarding national interests in 1993. She is an IEEE Computer Society Golden Core Member, and received the IEEE Millennium Medallion. A 2004 YWCA Women on the Move recipient for the Rio Grande Chapter, Dr. Pollock also received recognition from the 46th New Mexico State Legislature for her community activities. She received the Texas A&M University Distinguished Former Student Award in 2006.
CSCE 681 Contact: Krista Simmons (krista [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Researching Engineering Education: Some Philosophical Considerations
Dr. Sandra Cairncross
Dean & Senior Teaching Fellow
Faculty of Engineering Computing & Creative Industries
Edinburgh Napier University
4:10 p.m., Monday October 12, 2009
Room 124, Bright Building
Abstract
Enhancements to teaching, learning and assessment in engineering require the underpinning of research and scholarship into the academic practice of engineering education as well as that of disciplined based research, if they are to lead to lasting benefits for students. Research into engineering education can take different forms ranging from large scale multi-institutional studies to cross-institutional impact analysis to individual teachers undertaking action research and can often be challenging, requiring an understanding of not only how students learn but also an understanding of educational research methods. There are a variety of different and often contradictory approaches to educational research which can make it difficult for the novice researcher to decide which one to select for a given study. Indeed many academics, when first embarking on research into education, naturally draw upon the research methods associated with their own discipline. However the most appropriate method for researching a given subject is not necessarily the most appropriate for researching the education of that subject. A case study is presented on the experiences of one researcher into engineering education research, as they diversified from methods rooted in their own discipline into methods borrowed from other disciplines. Consideration of underlying philosophical concepts frames this journey in a wider context, allowing key concerns to be explored. These concerns include the value of controlled experiments in the context of engineering education and the use of interpretativist approaches to illuminate particular situations. The importance of understanding different methodologies, and the claims that they can support, will be discussed, both in the context of conducting one's own research but also in the context of interpreting the work of others, thereby enabling researchers to take a more critical approach. This is of particular importance when seeking to apply findings of others to one's own practice as well as placing one's own research in an appropriate theoretical context. Ethical issues are also considered. Educational research should be prompted by a desire to improve the common good and should seek to minimise harm to learners. Ethical considerations should inform both the conduct of the research and the dissemination of findings. This can prove problematic when undertaking practitioner-based research. Key aspects will be explored. These reflections will be useful to others embarking on their own research projects. Drawing upon key western philosophical traditions, social science theories and pedagogy, the paper argues that an understanding of key aspects of philosophy, in particular consideration of "what is knowledge?" and "how we come to know" can improve the design of investigations into how students learn, and how that learning can be enhanced.
Biography
Dr Sandra Cairncross was appointed as Dean of Engineering, Computing & Creative Industries in April 2008. Prior to that she was Associate Dean with responsibility for Academic Quality and Customer Service. Her role is to lead the strategic development of the Faculty, building on its excellent track record in supporting Scotland's knowledge based economy through providing a portfolio of academic programmes and engaging in research and knowledge transfer which are relevant to the needs of students, business and industry and other stakeholders.
Dr Cairncross is Chartered Engineer and a member of the Institute of Educational Technology and a Fellow of the Higher Education Academy with a background in Interactive Media Design. She is a Senior Teaching Fellow at Edinburgh Napier and the focus of her doctoral students was how best to use new learning technologies to enhance teaching and learning and the student experience and has published in this area.
CSCE 681 Faculty Contact: Dr. Tracy Hammond (hammond [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Adventures in Electronic Voting Research
Dr. Dan Wallach
Associate Professor
Department of Computer Science
Rice University
4:10 p.m., Monday October 19, 2009
Room 124, Bright Building
Abstract
In elections employing electronic voting machines, we have observed that poor procedures, equipment failures, and honest mistakes pose a real threat to the accuracy of the final tally. The event logs kept by these machines can give auditors clues as to the causes of anomalies and inconsistencies; however, each voting machine is trusted to keep its own audit and ballot data, making the record unreliable. If a machine is damaged, accidentally erased, or otherwise compromised during the election, we have no way to detect tampering or loss of auditing records and cast votes.
This talk begins with our experiences in real elections where we have observed these issues in the field, including a disputed primary election in Laredo, Texas as well as the recent Congressional election in Sarasota, Florida. These issues motivate a new design for a voting architecture we call "VoteBox" which networks the voting machines in a polling place, allowing for replicated, timeline-entangled logs which can survive malice and malfunction to provide a verifiable audit of election-day events.
Biography
Dan Wallach is an associate professor in the Department of Computer Science at Rice University in Houston, Texas and is the associate director of NSF's ACCURATE (A Center for Correct, Usable, Reliable, Auditable and Transparent Elections). His research involves computer security and the issues of building secure and robust software systems for the Internet. He has testified about voting security issues before government bodies in the U.S., Mexico, and the European Union, has served as an expert witness in a number of voting technology lawsuits, and recently participated in California's "top-to-bottom" audit of its voting systems.
CSCE 681 Faculty Contact: Dr. Guofei Gu (guofei [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Physics-Based Animation and Modeling: Spider Web, Raindrop Behavior, and Polyhedra
Dr. Yonggao Yang
Associate Professor
Department of Computer Science
Prairie View A&M University
4:10 p.m., Monday October 26, 2009
Room 124, Bright Building
Abstract
Physics-based animation applies physics laws to govern animation. It allows us to achieve high quality and realistic natural phenomena animation that has high demand in video games and animated movies. This talk focuses on using physics-based animation to animate several natural phenomena, including spider web and raindrop behavior on a glass window, and then extending its use to construct polyhedra. The speaker will also briefly introduce other projects he worked or is working on, including subdivision-based geometry design, VRMath - using virtual reality to enhance college math teaching and learning, knee model construction, and others.
Biography
Yonggao Yang is an associate professor and interim department head in the Computer Science Department at Prairie View A&M University. He received his B.S. and M.S. in Computer Science from Southwest Jiaotong University (China) in 1984 and 1987, respectively, and his Ph.D. in Information Technology from George Mason University in 2002. He joined Prairie View A&M University in 2002. His research involves Computer Graphics, Virtual Learning/Training Environments, Scientific Visualization, and Computer Network Security.
CSCE 681 Faculty Contact: Dr. Nancy Amato (amato [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Can We Really Use 1.5 Million Cores Effectively?
Dr. Bronis de Supinski
Advanced Simulation and Computing
Lawrence Livermore National Laboratory (LLNL)
4:10 p.m., Wednesday October 28, 2009
Room 124, Bright Building
Abstract
We are faced with an explosion in parallelism at all levels of large scale systems. Multi-core chips have become ubiquitous and almost all large scale systems use them. Systems such as Lawrence Livermore National Laboratory's BlueGene/L and Oak Ridge National Laboratory's Jaguar already have over 100,000 processor cores. The recently announced ASC Sequoia will have over 1.5 million cores when it is deployed in FY12. Los Alamos National Laboratory's Roadrunner system features a heterogeneous node architecture that requires the use of three different compilers to build a single application.
Hardware support for other novel parallelism mechanisms, such as transactional memory and thread level speculation, are likely to appear in systems in the near future. Further, future systems are likely to have much less off-chip and off-node bandwidth per core as well as significantly smaller main memories per core. These trends will necessitate significant changes in applications and the development environment that supports them. We will require new mechanisms to target applications to these architectures, to identify and to solve software defects that arise in those applications and to understand and to improve their performance.
In this talk, I will detail the overall National Nuclear Security Administration's Advanced Simulation and Computing (ASC) Program's development environment strategy for current and future large scale systems and several novel directions that we are pursuing as part of that strategy.
Biography
Bronis R. de Supinski is the co-leader in the Advanced Simulation and Computing (ASC) program's Application Development Environment and Performance Team (ADEPT) at Lawrence Livermore National Laboratory (LLNL). His research interests include high performance computer architectures, performance modeling and analysis, message passing implementations and tools, memory performance improvement, cache coherence and distributed shared memory, consistency semantics and programming models. Bronis earned his Ph.D. in Computer Science from the University of Virginia in 1998 and he joined LLNL's Center for Applied Scientific Compuiting (CASC) in July 1998. Currently, his projects include applications of data mining techniques to performance analysis and modeling, scalable debugging methods, investigations into mechanisms and tools to improve memory performance, a variety of optimization techniques and tools for MPI and several issues with OpenMP, including its memory model and tool support. He pursues the last set of topics as the Chair of the OpenMP Language Committee. Throughout his career, Bronis has won several awards, including the prestigious Gordon Bell Prize in 2005 and 2006. He is a member of the ACM and the IEEE Computer Society.
CSCE 681 Faculty Contact: Dr. Nancy Amato (amato [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Achieving Timeliness Without Clocks
Peter Robinson
Embedded Computing Systems Group
Institute of Computer Engineering
Vienna University of Technology
4:10 p.m., Monday November 2, 2009
Room 124, Bright Building
Abstract
In this talk, we will show how synchrony conditions can be added to the asynchronous distributed computing model in a way that avoids any reference to message delays and computing step times, as well as any global constraints on communication patterns and network topology. More specifically, we will introduce the Asynchronous Bounded-Cycle (ABC) model, where clock synchronization and lock-step rounds can easily be implemented and proved correct, even in the presence of Byzantine failures. Furthermore, we will describe some weaker variants of the ABC model that allow even more relaxed system assumptions. We will also see how the ABC model relates to existing partially synchronous system models, in particular, to the classic models of Dwork, Lynch and Stockmayer, and discuss some aspects of the ABC model's applicability in real systems.
Biography
Peter Robinson is a Research Assistant at the Vienna University of Technology (Austria). He received his M.S. and B.S. degrees in computer science from the Vienna University of Technology where he is currently working on his Ph.D. His research interests focus on fault-tolerant distributed computing, and, more specifically, on system models and algorithms with weak synchrony requirements.
CSCE 681 Faculty Contact: Dr. Jennifer Welch (welch [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Linking Documents to Encyclopedic Knowledge: Using Wikipedia as a Source of Linguistic Evidence
Dr. Rada Mihalcea
Associate Professor
Department of Computer Science and Engineering
University of North Texas
4:10 p.m., Monday November 16, 2009
Room 124, Bright Building
Abstract
Wikipedia is an online encyclopedia that has grown to become one of the largest online repositories of encyclopedic knowledge, with millions of articles available for a large number of languages. In fact, Wikipedia editions are available for more than 200 languages, with a number of entries varying from a few pages to more than one million articles per language.
In this talk, I will describe the use of Wikipedia as a source of linguistic evidence for natural language processing tasks. In particular, I will show how this online encyclopedia can be used to achieve state-of-the-art results on two text processing tasks: automatic keyword extraction and word sense disambiguation. I will also show how the two methods can be combined into a system able to automatically enrich a text with links to encyclopedic knowledge. Given an input document, the system identifies the important concepts in the text and automatically links these concepts to the corresponding Wikipedia pages. Evaluations of the system showed that the automatic annotations are reliable and hardly distinguishable from manual annotations. Additionally, an evaluation of the system in an educational environment showed that the availability of encyclopedic knowledge within easy reach of a learner can improve both the quality of the knowledge acquired and the time needed to obtain such knowledge.
Biography
Rada Mihalcea is an Associate Professor in the Department of Computer Science and Engineering. Her research interests are in computational linguistics, with a focus on lexical semantics, graph-based algorithms for natural language processing, and multilingual natural language processing. She is currently involved in a number of research projects, including word sense disambiguation, monolingual and cross-lingual semantic similarity, automatic keyword extraction and text summarization, emotion and sentiment analysis, and computational humor. She serves or has served on the editorial boards of the Journals of Computational Linguistics, Language Resources and Evaluations, Natural Language Engineering, and Research in Language in Computation. Her research has been funded by the National Science Foundation, Google, and the State of Texas. She is the recipient of a National Science Foundation CAREER award (2008) and a Presidential Early Career Award for Scientists and Engineers (PECASE, 2009).
CSCE 681 Faculty Contact: Dr. Andruid Kerne (andruid [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Temporal Debugging using URDB
Gene Cooperman
Institute for Complex Scientific Software (ICSS)
Northeastern University
4:10 p.m., Wednesday November 18, 2009
Room 124, Bright Building
Abstract
This talk presents URDB, a universal reversible debugger. URDB supports reversible debugging for: gdb, MATLAB, python, and perl. Reversibility support for an additional debugger can be added in a few hours. In addition to adding standard reversible commands (for example, reverse-next, reverse-step, reverse-continue, reverse-finish), URDB adds temporal search. For example, it uses binary search through a process lifetime to bring a debugging session to exactly the program statement whose execution will cause a user-defined expression to change its value.
URDB supports temporal debugging both on multi-threaded SMP architectures and on multi-process architectures over distributed nodes. URDB gains its temporal feature through the use of DMTCP (Distributed MultiThreaded CheckPpointing), a fast checkpoint-restart package.
URDB has been under development for more than a year. This talk will describe the history of URDB and its DMTCP antecedent, the experience with URDB to date, and a vision for the future of temporal debugging and other temporal support tools.
Biography
Gene Cooperman received his Ph.D. from Brown University in 1978. He spent two years as a post-doc, followed by six years at GTE Laboratories. Since 1986, he has been a professor at Northeastern University. His interests lie in high performance computation and symbolic algebra. The combination of these two subjects has also led to his joint work with his students on a second theme: disk-based parallel computation. He leads the High Performance Computing Laboratory at Northeastern University, where he currently advises five PhD students. He has over 80 refereed publications.
CSCE 681 Faculty Contact: Dr. Gabriel Dos Reis (gdr [at] cse.tamu.edu)
CSCE 681 Graduate Seminar:
Energy in the 21st Century: building a brighter future!
Dr. Najib Abusalbi
Director of University Collaboration
Schlumberger
4:10 p.m., Wednesday December 2, 2009
Room 124, Bright Building
Abstract
How long will oil and gas reserves last. and does it matter? Many economists and scientists have made predictions, explored options and written about their theories. Some predict a World War III triggered by competition for oil in ten years, others talk about "bottomless" reserves. And some people say that if we just turn lights off at night, and drive hybrid cards, we'll be fine. Beyond the hype, what are the facts? This talk will cover where energy comes from and where it is consumed, how this balance has been evolving, and what seems to be in store for the world in the next decades, with focus on the market, the technology and the people that will shape this future.
Biography
Najib holds a PhD. in Atomic Physics from Louisiana State University. He joined Schlumberger in 1984 after three years in postdoctoral research and teaching assignments in Physics and Chemistry. Since then, he held multiple product development and management positions in several exploration and production domains – including information management director, regional operations manager, project director and program manager, as well as recruiting, training and career development manager, innovation and research director, and most recently chief architect for production & operations software in upstream oil & gas. Najib Abusalbi is currently the Director of University Collaboration and is recognized as a Technology Advisor in Schlumberger's Communities of Practice; among his many contributions, he leads the Project Management community since 2005 and acts as a mentor.
CSCE 681 Faculty Contact: Dr. Nancy Amato (amato [at] cse.tamu.edu)
