USC - Viterbi School of Engineering

Apr
01

CS Colloq: Securing the Web With Decentralized Information Flow Control
Tue, Apr 01, 2008 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars

Title: Securing the Web With Decentralized Information Flow ControlSpeaker: Maxwell Krohn (MIT)Abstract:
The recent successes of server-side applications (e.g. Google and Facebook applications) hint that tomorrow's computing platform might not be the local desktop but rather the extensible remote Web site. Unfortunately, these new server-side platforms, built on conventional operating systems, are committing the same security mistakes already ossified in today's insecure desktops. In this talk, I will discuss how to secure both today's Web sites and tomorrow's Web computing platforms with a new OS technique called Decentralized Information Flow Control (DIFC). A DIFC system tracks the flow of secret data as it is copied from file to file and communicated from process to process. In the end, the OS lets modules known as "declassifiers" legislate policies for secret data exiting to the network. DIFC provides better security than standard OSes because it allows developers to concentrate security-critical code in small, audit-friendly declassifiers, which remain small and contained even as the overall system balloons with new features. This talk presents DIFC, an implementation of DIFC for Linux, and a case study of a complex, popular open-source application (MoinMoin Wiki) secured with DIFC. MoinMoin is a prototype for more ambitious and general work to come, such as a novel Web-based application platform with encouraging security guarantees. Joint work with: Micah Brodsky, Natan Cliffer, Petros Efstathopoulos, Cliff Frey, Eddie Kohler, David Mazieres, Robert Morris, Frans Kaashoek, Steve VanDeBogart, Mike Walfish, Alex Yip, David ZieglerBiography:
Maxwell Krohn is a PhD candidate in Computer Science at MIT. He received his BA from Harvard in 1999, and was a staff research scientist at NYU from 2002-2003. In between, he co-founded and co-built several community Web sites, some vintage (TheSpark.com), others live and kicking (SparkNotes.com and OkCupid.com). His research interests are in operating systems, distributed systems and security.
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited

Contact: CS Colloquia
Apr
02

Length and Time Scales of Trapping Processes Associated with CO2 Sequestration
Wed, Apr 02, 2008 @ 02:00 PM - 03:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars

Speaker: Dr. Hamdi Tchelepi, Stanford University Abstract:
We study the primary CO2 trapping processes in deep saline aquifers:
dissolution and residual trapping. The length and time scales associated with the onset of miscible instability and the long term convective dissolution rates are analyzed using linear stability analysis and high-resolution nonlinear simulations. The migration of CO2 plumes as gravity currents and the effect of residual trapping of the CO2 in the wake of the advancing current are analyzed using a one-dimensional sharp-interface model. We study the behavior of gravity currents in large-scale aquifers with small dip angles, and we discuss issue related to propagation speeds, trapping capacities, and maximum plume migration distances. These preliminary results indicate that CO2 storage in large, deep saline aquifers may be a viable option.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited

Contact: Evangeline Reyes
Apr
02

Electronically tunable nanomaterials
Wed, Apr 02, 2008 @ 03:30 PM - 04:30 PM
Aerospace and Mechanical Engineering
Conferences, Lectures, & Seminars

Horst HahnForschungszentrum KarlsruheInstitute for Nanotechnology and Joint ResearchLaboratory NanomaterialsTechnische Universität DarmstadtAbstract:The properties of materials are typically controlled in a static manner by the microstructure. This implies control of the grain size, defect concentration, structure and metastability. As long as the microstructure does not change during the use of the material, the properties of the material are fixed, or irreversible. In contrast, in semiconducting materials, properties can be tuned by the application of an external field due to the space charge regions which extend far from the interfaces. In metallic systems, this effect cannot be observed unless the dimensions of the structures are in the nanometer regime. The reason for this different behaviour is the small spatial dimension of the space charge regions due to the effective screening of the induced charges by the conduction electrons.
In nanoporous metals and thin films exposed to appropriate electrolytes, it has been demonstrated that substantial changes of physical properties can be induced by the application of a potential between the nanostructured metal and a counter electrode. Examples of the changes of surface stresses and the electrical resistivity of thin Gold films and nanoporous Gold will be presented. A simple model is proposed based on the modification of the electron density distribution at the interface of the metal and the electrolyte. Effectively, the corresponding change of the effective thickness of the sample is the major cause of the observed resistivity change.
Additionally, a transparent conducting oxide, ITO, in a nanoparticulate form has been prepared from a dispersion using spin coating. The observed resistivity changes, i.e. the on/off ration can be as large as 2.000, i.e. 200.000 %, between the different values of the control potential. Moreover, the device exhibits field effect transistor behavior identical to a conventional semiconductor, but in this case observed in a material with a large charge carrier density exhibiting metallic conduction behaviour. Additionally, the mobility is exceeding 20 cm2/Vs. The device can be used for printable electronics and transparent electronics.

Location: Seaver Science Library, Rm 150
Audiences: Everyone Is Invited

Contact: April Mundy
Apr
02

CS Colloq: Acquisition, Compression, and Transfer of Reflectance Fields
Wed, Apr 02, 2008 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars

Title: Acquisition, Compression, and Transfer of Reflectance FieldsSpeaker: Dr. Pieter Peers (ICT)Abstract:
A reflectance field describes the light transport through a scene in terms of incident and radiant illumination. Full knowledge of a scene's reflectance field allows to view this scene from any viewpoint and under any illumination condition. This simple yet powerful formulation is the basis for many image-based methods, and is the main focus of my research. My research can be categorized into three main topics: acquisition, compression and transfer of reflectance fields. For each of these research topics, a selected contribution is discussed in detail. The first presented contribution describes an acquisition method that enables to capture detailed reflectance fields for image-based relighting using non-adaptive illumination (i.e., measurement) patterns. These measurement patterns are incoherent with the reflectance field itself, and the number of measurement patterns is proportional to the size of the compressed field, rather than the size of full uncompressed field. The second discussed contribution details a novel compression method for (measured) heterogeneous subsurface scattering, i.e., the way light scatters through a non-homogeneous semi-opaque medium. This compression is based on a specially developed matrix factorization method. A third contribution, demonstrates a post-production method that successfully generates plausible relit sequences of dynamic facial performances of a subject. Relighting is achieved by transferring reflectance information from a single reflectance field of a static pose of a potentially different subject, but similar in appearance. In the final part of this presentation, an overview of short term and long term research plans are given. Additionally, some preliminary results of recent research are shown.Biography:
Pieter Peers is currently a senior researcher in the Graphics Lab at the Institute for Creative Technologies (ICT) of the University of Southern California (USC). Before that he was a research assistant in the Computer Graphics Research Group at the K.U.Leuven (Belgium), where he also obtained his Ph.D. in August 2006. His research focuses on data-driven computer graphics, in particular image-based relighting.
Location: Henry Salvatori Computer Science Center (SAL) - 322
Audiences: Everyone Is Invited

Contact: CS Colloquia
Apr
03

Making Fuel and Drug-Producing Microbes through Analysis, Modeling and Design
Thu, Apr 03, 2008 @ 10:30 AM - 11:30 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

Speaker: Desmond Lun, Ph.D.Abstract:Engineering microorganisms that efficiently produce drugs and fuels is an exciting and challenging problem
with large potential impact on energy supply, the environment, and global health. Such engineering is greatly
aided by systematic design and, in this talk, we discuss how systematic design can be achieved through the
analysis and modeling of microbial metabolic networks. We discuss approaches that we are developing for
modeling metabolism and gene regulation and for using these models to guide design. In particular, we
describe a network optimization problem that arises in the context of optimal design and discuss algorithmic
approaches for its solution. We describe our progress in engineering E. coli for petroleum production from
simple sugars and in other engineering directions.Biography:Desmond Lun is a Computational Biologist at the Broad Institute of MIT and Harvard and a Research Fellow in
Genetics at Harvard Medical School. Prior to his present position, he was a Postdoctoral Research Associate in
the Coordinated Science Laboratory at the University of Illinois at Urbana-Champaign. He received bachelor's
degrees in mathematics and computer engineering from the University of Melbourne, Australia in 2001, and
S.M. and Ph.D. degrees in electrical engineering and computer science from MIT in 2002 and 2006,
respectively. Dr. Lun's research interests are in synthetic biology, systems biology, and networking. He is coauthor,
with Tracey Ho, of "Network Coding: An Introduction," forthcoming from Cambridge University Press.Host: Professor C.-C. Jay Kuo
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited

Contact: Talyia Veal
Apr
03

Interference Channels with Generalized Feedback
Thu, Apr 03, 2008 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

SPEAKER: Professor Daniela Tuninetti, ECE Department, University of Illinois-ChicagoABSTRACT: In the past years, there has been a renewed interest to characterize the ultimate performance of InterFerence Channels (IFC). Classical IFCs model competition among uncoordinated source-destination pairs. IFCs with generalized feedback model scenarios where the sources can sense the current channel activity, like in wireless channels, and use this information to communicate cooperatively. Although cooperative communications is not equivalent to virtual MIMO communications, it has been shown that it benefits the performance of all the involved source-destination pairs without increasing neither the transmit powers nor the channel bandwidth. In this talk, we will review recent progresses for IFCs with and without generalized feedback. We will describe the currently best achievable and outer-bound regions. We will conclude by comparing those regions for Gaussian IFCs under different feedback configurations, and pointing out open problems.HOST: Prof. Giuseppe Caire, caire@usc.edu
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited

Contact: Mayumi Thrasher
Apr
03

Peptide Materials Engineering
Thu, Apr 03, 2008 @ 03:30 PM
Mork Family Department of Chemical Engineering and Materials Science
Conferences, Lectures, & Seminars

Announcing The USC Inaugural Pings LectureshipwithProfessor Matthew TirrellDepartments of Chemical Engineering and Materials
Materials Research Laboratory
Institute for Collaborative Biotechnologies
California NanoSystems Institute
University of California, Santa Barbara
Santa Barbara, CA 93106-5130
Tel:(805) 893-3141, Fax:(805) 893-8124
E-mail: tirrell@engineering.ucsb.eduAbstractPeptides are functional modules of protein macromolecules that can be
displayed apart from the whole protein to create biofunctional surfaces
and interfaces, or can be re-assembled in new ways to create synthetic mimics of protein structures. Each of these routes are being employed to gain new insight into protein folding and to develop new, functional, biomolecular materials. Examples of work from our laboratory in this area using peptide-lipid conjugate molecules (peptide amphiphiles) will be discussed relating to multi-functional surfaces, liposomal drug delivery, protein analogous micelles, DNA-binding peptide modules and anti-microbial peptides.Professor Matthew Tirrell is Dean of Engineering at UC, Santa Barbara. He received a B.S. in Chemical Engineering at Northwestern University and a Ph.D. in 1977 in Polymer Science from the University of Massachusetts. From 1977 to 1999 he was on the faculty of Chemical Engineering and Materials Science at the University of Minnesota, where he served as head of department from 1995 to 1999. His research has been in polymer surface properties, adsorption, adhesion, surface treatment, friction, lubrication and biocompatibility. He has co-authored about 270 papers and one book and has supervised about 70 Ph.D. students. Professor Tirrell has been a Sloan and a Guggenheim Fellow, a recipient of the Camille and Henry Dreyfus Teacher-Scholar Award and has received the Allan P. Colburn, Charles Stine, William H. Walker and the Professional Progress Awards from AIChE, and was the Institute Lecturer in 2001. He is a member of the National Academy of Engineering, and a Fellow of: the American Institute of Medical and Biological Engineers, the AAAS, and the APS. In 2003, he concluded two years of service as co-chair of the steering committee for the National Research Council: "Beyond the Molecular Frontier: Challenges for Chemistry and Chemical Engineering".

Location: Henry Salvatori Computer Science Center (SAL) - 101
Audiences: Everyone Is Invited

Contact: Petra Pearce Sapir
Apr
03

CS Colloq: Places Everyone: Creating an Animated Tapestry of Human Activity for Virtual Worlds
Thu, Apr 03, 2008 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars

Title: Places Everyone: Creating an Animated Tapestry of Human Activity for Virtual WorldsSpeaker: Jan Allbeck(UPENN)Abstract:
As we journey through our day, our lives intersect with other people. We see people leaving for work, waiting for trains, meeting with friends, hard at work, and thousands of other activities that we may not even be conscious of. People create a rich tapestry of activity throughout our day, a human texture. We may not always be aware of this texture, but we would definitely notice if it were missing, and it is missing from many simulations. Creating virtual scenarios that simulate a substantial human population with typical and varied behaviors can be an overwhelming task. In addition to modeling the environment and characters, tagging the environment with semantic data, and creating motions for the characters, the simulation engineer also needs to create character profiles for a heterogeneous population and link these character traits to appropriate behaviors to be performed at appropriate times and in appropriate places during the simulation. Due to the large number of individuals, the variety of behaviors they may engage in, and the potential complexity of environments, this is currently beyond the scope of military, crowd research, or entertainment simulations. At present, simulations either have a very limited number of character profiles or are meticulously hand scripted. I will describe a framework, called CAROSA (Crowds with Aleatoric, Reactive, Opportunistic, and Scheduled Actions), that will facilitate the creation of heterogeneous populations for large scale simulations by using a commercial off-the-shelf software package (Microsoft Outlook®), a Parameterized Action Representation (PAR), and multiple human agent simulation software (HiDAC). CAROSA incorporates four different broad action types: scheduled, reactive, opportunistic, and aleatoric. Scheduled activities arise from specified roles for individuals or groups; reactive actions are triggered by contextual events or environmental constraints; opportunistic actions arise from explicit goals and priorities; aleatoric actions are random but structured by choices, distributions, or parametric variations. The CAROSA architecture enables the specification and control of actions for more realistic large scale human textures in virtual worlds such as buildings and cities, links human characteristics and high level behaviors to animated graphical depictions, and relieves some of the burden in creating and animating heterogeneous 3D animated human populations.Biography:
I am a Ph.D. candidate in the Department of Computer and Information Science, which is a part of the School of Engineering and Applied Science at the University of Pennsylvania. My advisor is Dr. Norman I. Badler. I am also Associate Director of the Center for Human Modeling and Simulation (HMS), where I coordinate and participate in the research projects affiliated with HMS as well as coordinating the operational aspects of the lab facility. I have Bachelors degrees in Mathematics and Computer Science from Bloomsburg University and a Masters degree in Computer and Information Science from Penn. I have had the great opportunity to explore many aspects of computer graphics, but am most drawn to research at the crossroads of animation, artificial intelligence, and psychology in the simulation of virtual humans. My current research focuses on the creation and simulation of heterogeneous, functional crowds.
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited

Contact: CS Colloquia
Apr
04

Going Beyond Diffusion Tensor Imaging: Local Reconstruction and Tractography in High Angular..
Fri, Apr 04, 2008 @ 10:30 AM - 11:30 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

Speaker: Maxime Descoteaux, Ph.D. NMR Lab Neurospin / CEA Saclay
FRANCE Abstract:
At the current resolution of diffusion-weighted (DW) magnetic resonance imaging (MRI), research groups agree that there are between one third to two thirds of imaging voxels in the human brain white matter that contain multiple fiber bundles crossing. This presentation tackles the important problem of recovering crossing fiber bundles from high angular resolution diffusion imaging (HARDI) data. The main goal is to overcome the limitations of diffusion tensor imaging (DTI). It is well-known that imaging voxels where there are multiple fiber crossings are locations where DTI is limited and inadequate. In this presentation, a simple, fast and robust Q-ball imaging (QBI) reconstruction is presented using spherical harmonics. QBI is a recent HARDI technique that reconstructs the orientation distribution function (ODF) of the average diffusion of the water molecules in the underlying fiber population. It is able to describe multiple fiber populations crossing. From this diffusion ODF, we describe how we can reconstruct the fiber ODF in order to perform accurate tractography. We develop a new spherical deconvolution sharpening method that transforms the diffusion ODF into a fiber ODF. Finally, we develop a new deterministic tractography algorithm and a new probabilistic tractography algorithm exploiting the full multi-directional information of the fiber ODF. We show local reconstructions and tracking results on complex fiber regions with known fiber crossings from simulated HARDI data, from a biological phantom and from multiple human brain datasets. Most current DTI based methods neglect these complex fibers, which might lead to wrong interpretations of the brain anatomy and functioning.Host: Professor Richard Leahy
Location: Hedco Neurosciences Building (HNB) - 100
Audiences: Everyone Is Invited

Contact: Talyia Veal
Apr
04

EE Students Practical Guide Seminar Series - Picking a Research Topic
Fri, Apr 04, 2008 @ 11:30 AM - 01:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

Seminar Leaders: Profs. Konstantinos Psounis and Alan WillnerWebsite: http://ee.usc.edu/news/practical-guide/* Pizza will be provided by the EE Department.*Abstract: Selecting a research topic is important. It sets the course for the next three-four years of a student's life. What is more, it will define the area of a PhD's job search, and it my very well define the area of interest of a researcher in the industry or academia for years to come. The right research topic can set you on a successful path, and choosing wisely can be a combination of insight and luck. Alternatively, choosing the wrong research topic can cause major problems for any graduate student. Following some introductory remarks regarding right ways and wrong ways to approach this critical problem, this will be an informal conversational meeting with Dr. Psounis and Dr. Lidar, where students will be free to ask any questions about how to choose a good research topic.Biographical SketchesProf. Konstantinos Psounis is an assistant professor of Electrical Engineering and Computer Science at the University of Southern California. He received his first degree from the department of Electrical and Computer Engineering of National Technical University of Athens, Greece, in June 1997, the M.S. degree in Electrical Engineering from Stanford University, California, in January 1999, and the Ph.D. degree in Electrical Engineering from Stanford University in December 2002. Konstantinos models and analyzes the performance of a variety of networks, including the Internet, mobile ad hoc networks, delay and disruptive tolerant networks, sensor networks, mesh networks, peer to peer networks and the web. He also designs methods and algorithms to solve problems related to such systems. He is the author of more than 40 research papers on these topics. Konstantinos has received faculty awards from NSF and the Zumberge foundation, has been a Stanford graduate fellow throughout his graduate studies, and has received the best-student National Technical University of Athens award for graduating first in his class.
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited

Contact: Gerrielyn Ramos
Apr
04

Codes and Capacity for Bidirectional Communication on Lines, Stars, and Rings
Fri, Apr 04, 2008 @ 01:00 PM - 02:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

SPEAKER: Dr. Gerhard Kramer, Member of Technical Staff, Communications and Statistical Sciences Dept., Bell Labs, Alcatel-LucentABSTRACT: This talk describes codes and capacity bounds for two-way communication in three classes of networks that are elements of larger networks. We begin by considering bidirectional line networks with both edge and node capacity constraints, as well as multiple multi-cast traffic sessions. For wired line networks, we construct an explicit binary network code that achieves capacity. The converse theorem requires using our progressive d-separating edge set (PdE) bound, which is a fundamental generalization of a standard cut bound. Second, we extend the results to line networks with broadcasting nodes by studying a two-way relay channel with three nodes: two sources and a relay. We develop several coded modulation techniques that achieve all rate points inside the capacity region. Third, we extend coding for the three-node topology to many nodes and develop a separation-based coding method that achieves good rates. We show that this approach achieves capacity if the broadcast channels are physically degraded and the multi-access channels exhibit uniform-phase fading. Finally, we outline extensions of the ideas to star and ring networks.BIO: Gerhard Kramer received the B.Sc. and M.Sc. degrees in electrical engineering from the University of Manitoba in Canada and the Dr. sc. techn. (Doktor der Technischen Wissenschaften) degree from the Swiss Federal Institute of Technology (ETH) in Zurich, Switzerland. From July 1998 to March 2000, he was with Endora Tech AG, Basel, Switzerland, as a communications engineering consultant. Since May 2000 he has been with Bell Laboratories, Murray Hill, NJ, USA. His research has been focused on information theory, communications theory, iterative decoding, and source coding.Dr. Kramer is currently serving as an Associate Editor for Shannon Theory for the IEEE Transactions on Information Theory. He served as Guest Editor-in-Chief for an IEEE Transactions on Information Theory Special Issue on Relaying and Cooperation in Communication Networks in 2006-2007, and as Publications Editor for the same Transactions during 2004-2005. He is serving as Co-Chair of the Technical Program Committee for the 2008 IEEE International Symposium on Information Theory. Dr. Kramer is a co-recipient of the IEEE Communications Society 2005 Stephen O. Rice Prize paper award, a Bell Labs President's Gold Award in 2003, and a recipient of an ETH Medal in 1998.HOST: Dr. Alan Willner, willner@usc.edu
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited

Contact: Mayumi Thrasher
Apr
04

Interm-scale Testing for Process Understanding, Model Validation & Up-scaling of Flow &Trans
Fri, Apr 04, 2008 @ 01:00 PM - 02:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars

Tissa Illangasekare, PE, DEE, DWREAMAX Distinguished Chair of Environmental Sciences and Engineering
And Professor of Civil Engineering
Director, Center for Experimental Study of Subsurface Environmental Processes (CESEP)
Colorado School of Mines, Golden, Colorado, USAABSTRACTGeologic heterogeneity plays a significant role in water flow and complex behavior of chemicals and waste products in the subsurface. A complete knowledge of the governing processes and how they are affected by the heterogeneity are difficult to obtain at field sites due to cost constraints and limitations of currently available technologies and methods for subsurface characterization. Characterization data limitations and lack of access and control at field sites make it difficult to validate theories and prediction models simulating complex flow and transport processes. Intermediate-scale, physical models provide cost effective alternatives that allows for the generation of accurate and high-resolution data at a range of observational scales, under controlled conditions in synthetically created aquifers that are highly instrumented for automated data acquisition. Examples involving numerical and conceptual model validation, soft and hard data assimilation in model calibration, evaluation of remediation technologies, up-scaling from laboratory to field systems and development of new sensor technologies for subsurface monitoring will be presented.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited

Contact: Evangeline Reyes
Apr
05

Grodins Graduate Research Symposium
Sat, Apr 05, 2008
Alfred E. Mann Department of Biomedical Engineering
Conferences, Lectures, & Seminars

http://bmsrs.usc.edu/~grodins/index.html
We are pleased to invite the graduate students of the Biomedical Engineering (BME) department to participate in the 12th Annual Grodins Graduate Research Symposium, to be held on Saturday, April 5, 2008 in USC's Galen Center. The symposium will include oral and poster presentations from almost all BME graduate students.
Location: Galen Center
Audiences: Graduate/Department/Sponsors only

Contact: Mischalgrace Diasanta
Apr
07

DICIT - An Example of Acoustic Scene Analysis and Distant-talking Based Spoken Dialogue System
Mon, Apr 07, 2008 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

Speaker: Maurizio OmologoAbstract:The aim of this talk is to introduce the DICIT EC project. DICIT focuses on the development of advanced
acoustic sensor based technologies and the related application to the smart-home environment. The main
foreseen application is an automatic voice interaction system, with users at a distance of up to some meters from
a set of microphones. Voice input operates in a complementary modality to the use of remote control, this way allowing one to control TV and related devices in a natural and flexible way. Due to the interference of other
coexisting active sound sources (e.g. loudspeakers, other talkers and noise sources) and to the effect of room
acoustics, both the processing and the understanding of given speech utterances become more problematic with
respect to the ideal situation tackled when using a close-talking microphone. To this regard, the talk will
provide an overview of the basic problems and technologies being developed as far as the multi-microphone
front-end are concerned. Other activities and results obtained during the first 18 months of the project will then be presented together with some video-clips showing the potential of the technologies so far developed.Bio:Maurizio Omologo was born in Padova (Italy) in 1959, and received the Laurea degree with honours in
Electrical Engineering (University of Padova) in 1984. From 1984 to 1987 he was Researcher at the Speech
Coding Department of CSELT - Torino (Italy). In 1988 he joined ITC-irst, as Senior Researcher of the Acoustic
and Speech Recognition group. Since 1998 he has been acting as head of the SHINE (Speech Interaction in
Noisy Environment) project at ITC-irst, now Bruno Kessler Foundation (FBK)-irst. From January 2003 to
December 2005 he was Associate Editor of the IEEE Transactions on Speech and Audio Processing journal. He
also has been teaching "Audio signal processing" at the Department of Information and Communication
Technology - University of Trento, since 2002. Currently, he acts as Coordinator of DICIT, a European Project
of the VI EC Framework Programme.Host: Prof. Shrikanth Narayanan
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited

Contact: Talyia Veal
Apr
07

Wen Li, Professor of Mechanical and Automation Engineering, Chinese University of Hong Kong
Mon, Apr 07, 2008 @ 12:30 PM - 01:50 PM
Alfred E. Mann Department of Biomedical Engineering
Conferences, Lectures, & Seminars

MEMS and Nanosensors
Audiences: Department Only

Contact: Mischalgrace Diasanta
Apr
07

Impulse-based UWB Systems and Applications
Mon, Apr 07, 2008 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

SPEAKER: Tor Sverre (Bassen) Lande, Professor, Dept. of Informatics, University of Oslo, NorwayABSTRACT: The old idea of impulse radio dates back to Marconi's first wireless transmissions using sparks. Unlike most wireless today, impulse radio transmissions are extremely wideband signals. The recently FCC-released frequency band from 3.1GHz to 10.6GHz is the widest unlicensed frequency band every released (7.5GHz). This ultra wide bandwidth (UWB) is commercially explored for even faster data transfer using traditional, multi-band (OFDM) RF techniques. However, the available bandwidth is wide enough for impulse radio transmission giving new functionality and new implementation challenges.I will show how power efficient impulse radio solutions are feasible in standard digital CMOS technology. Quite non-standard and untraditional design strategies must be used including time-domain signal processing. Circuit topologies for higher order Gaussian pulse generation and power efficient, correlating RAKE receivers will be explained.Impulse radio transmissions have additional interesting properties compared to narrowband modulation.-With time-domain processing (TDOA) highly accurate positioning is feasible in the millimeter range.-Improved sensitivity for robust communication.-Large number of channels (>100).Novel applications are also feasible using impulse transmission. Combining unconventional design techniques like "Swept-Threshold sampling" and digital lossless integration, micropower impulse radar is feasible in CMOS. A >30GHz sampler is used to accumulate and recover reflected electromagnetic energy. These new sensing devices (medical radar) may be explored for reading vital body signs (pulse, breathing) embedded in your car seat or hospital emergency bed. Just imagine what you can do looking though heavy matter!BIO: Tor Sverre Lande is a professor in the Microelectronic Systems at the Department of Informatics, University of Oslo. From 2004 he is also serving as a visiting professor at the Institute of Biomedical Engineering, Imperial College, London, UK. His primary research is related to microelectronics, both digital and analog. Research fields are neuromorphic engineering, analog signal processing, micropower circuit design, biomedical circuits and systems and impulse radio. He is the author or co-author of more than 90 scientific publications with chapters in two books. He is currently serving as an associate editor of several scientific journals. He has served as guest editor of special issues like IEEE Transactions on Circuits and Systems, vol. II, special issue on "Floating Gate Circuits and Systems", Jan 2001, and IEEE Transactions on Circuits and Systems, vol. I, special issue on "Biomedical Circuits and Systems", (Vol 52, 12, Dec. 2005). He is/has been a technical committee member of several international conferences and has served as reviewer for a number of international technical journals. He has served as Technical Program Chair for several international conferences (ISCAS 2003 in Bangkok, NORCHIP 2004 in Oslo, BioCAS workshop 2004 in Singapore, BioCAS 2006 in UK). He was chair elect (2003-2005) of the IEEE Biomedical Circuits and Systems technical committee (BioCAS) and is also a member of other CAS technical committee. In 2006 he was appointed Distinguished Lecturer of the IEEE Circuits and Systems Society (CAS) and elected member of CAS Board of Governors. He is also serving as Editor-in-Chief of the new IEEE Transactions on Biomedical Circuits and Systems. In 2006 he was appointed Senior Member of the IEEE.
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited

Contact: Mayumi Thrasher
Apr
08

Pseudospintronics for Ultra-Low Power Logic Devices
Tue, Apr 08, 2008 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

Matthew J. GilbertMicroelectronics Research Center, University of Texas at AustinAbstract
As the march towards ever smaller silicon devices continues unabated, we are rapidly approaching size scales where the bulk silicon transistor can no longer deliver sufficient device performance. The main problem with using charge based devices for next generation logic devices is that their performance has basic and fundamental physical limitations. While new device designs may extend the life of CMOS for several years, concerns about the power dissipation in these future generation CMOS devices has fueled the search for new computational state variables and the tools with which to evaluate these new devices.
In this talk, we will discuss the possibility of exploiting the exotic phenomena of strongly interacting systems to produce a completely new generation of logic devices based on collective behavior. In particular, we will discuss the device and transport properties of "pseudospintronic" systems. Pseudospintronics is a variant of spintronics where we represent the layer degree of freedom in a bilayer system (e.g. coupled quantum wells) as a spin. When the layers are separated by a small distance (~1 - 10 nm), the quasiparticles in each of the layers interact with one another. This interaction can greatly enhance the interlayer transport. We will discuss the application of this interaction-enhanced interlayer transport in III-V electron doped bilayers to elucidate the cause of the interlayer currents and the decay of the enhancement with increasing bias known as the "pseudospin torque effect". We also discuss a silicon based pseudospin system consisting of one electron doped layer and one hole doped layer. In this system we find reduced interlayer currents which arise from the discrepancy in dispersion relations between the two layers. Nevertheless, at elevated temperatures, both the III-V and silicon systems lose their interaction based enhancements.
We conclude our discussion by examining separately contacted, non-bonded graphene bilayer systems. These are devices that can have output characteristics very similar to a MOSFET, but while requiring much less switching energy. Furthermore, calculations show that the performance will not degrade at room temperature. Efforts are underway at Stanford University to experimentally realize graphene bilayer nanoswitches. Biography: Matthew Gilbert received the B.S. (Honors), M.S. and Ph.D. degrees from Arizona State University in 2000, 2003, and 2005 respectively all in electrical engineering. His Ph.D. research focused on novel systems for quantum computing and electron-phonon interactions in tri-gate nanowire transistors. He is currently the assistant director of the SouthWest Academy of Nanoelectronics (SWAN) and a post-doctoral fellow at the University of Texas at Austin. His research focuses on emergent semiconductor nanodevice technology which exploits computational state variables beyond that of charge (e.g. spin and phase) and their application to form beyond CMOS architectures. He has published over 40 conference and journal papers in the areas of spintronics, semiconductor nanowire MOS devices, graphene, computational algorithms for efficient transport calculation and correlated many-body systems and theory. Date: Tuesday, April 8, 2008
Place: OHE 120
Time: 2:00 PM – 3:00 PM

Location: Olin Hall of Engineering (OHE) - 120
Audiences: Everyone Is Invited

Contact: Ericka Lieberknecht
Apr
08

CS Colloq: Fluid Simulation With Reduced Diffusion, Dissipation, and Volume Loss
Tue, Apr 08, 2008 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars

Title: Fluid Simulation With Reduced Diffusion, Dissipation, and Volume LossSpeaker: Dr. ByungMoon Kim (GATECH)Abstract:
Recent advances in simulation methods for three-dimensional computer animation have led to a significant increase in realism and have benefited the digital entertainment industry. We will discuss two methods for improving the realism in fluid simulations: (1) The improved BFECC advection that increases the dynamic in simulated fluid motion and (2) a volume control technique that prevent the loss of fluid volume. We will demonstrate these advances on simulations of smoke, liquid, bubbles, rigid bodies, and foam. More detailed summaries of these two methods are provided below. We will also discuss future opportunities afforded by the symbiotic relation between scientific computing and computer animations. The BFECC (Back and Forth Error Compensation and Correction) was recently developed for interface computation using a level set method. We show that BFECC can be applied to reduce dissipation and diffusion encountered in a variety of advection steps, such as velocity, smoke density, and image advections on uniform and adaptive grids and on a triangulated surface. BFECC can be implemented trivially as a small modification of the first-order upwind or semi-Lagrangian integration of advection equations. It provides second-order accuracy in both space and time. When applied to level set evolution, BFECC reduces volume loss significantly. We demonstrate the benefits of this approach on image advection and on the simulation of smoke, bubbles in water, and the highly dynamic interaction between water, a solid, and air. We also apply BFECC to dye advection to visualize vector fields. Liquid and gas interactions often contain bubbles that stay for a long time without bursting on the surface, making a dry foam structure. Such long lasting bubbles simulated by the level set method can suffer from a slow but steady volume error that accumulates to a visible amount of volume change. We propose to address this problem by using the volume control method. We trace the volume change of each connected region, and apply a carefully computed divergence that compensates undesired volume changes. To compute the divergence, we construct a mathematical model of the volume change, choose control strategies that regulate the modeled volume error, and establish methods to compute the control gains that provide robust and fast reduction of the volume error, and (if desired) the control of how the volume changes over time.Biography:
ByungMoon Kim received a Ph. D. in computer science in 2006 at the Georgia Institute of Technology. At the same school, he received master's degrees in Aerospace Engineering in 1999, Computer Science in 2005, and Mathematics in 2005. He received a bachelor's degree in Aerospace Engineering in Inha University, Inchon, Korea in 1994. After receiving the Ph. D., he worked as a temporary professor in Mathematics at the Georgia Institute of Technology to teach a class and to research fluid simulation and image processing. After this, he joined NVIDIA Corp, where he worked on graphics device driver development, real time graphics research, and physics simulations. His research interests are in computer graphics, focusing on fluid simulation, geometry processing such as mesh filtering and editing, and haptic devices. He is an author of papers on various topics: mobile robot control, a spacecraft simulator, collision prediction, mesh editing, nonphotorealistic video processing, a mesh filter, a realtime shadow algorithm, and fluid simulations.
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited

Contact: CS Colloquia
Apr
09

Making Quantum Computers Fault Tolerant
Wed, Apr 09, 2008 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

SPEAKER: Dr. Ben Reichardt, Postdoctoral Fellow, California Institute of TechnologyABSTRACT: The biggest experimental obstacle to manipulating quantum information and realizing quantum computers is noise, or decoherence. Entangled quantum states are typically difficult to prepare without accumulating errors, and are highly susceptible to noise that collapses them down to merely classical states. General quantum fault-tolerance techniques, invented about a decade ago, can in theory solve both problems, but often require unrealistically low noise rates before they kick in.Over the last few years we have seen a renaissance in fault-tolerance schemes. These new schemes rely on quantum phenomena such as quantum teleportation to isolate the data from errors. I will describe these schemes that simulations indicate may tolerate as much as 3-6% noise per operation! However, as classical simulations of quantum systems are difficult, it is also important to develop rigorous methods to determine if, and how well, these schemes will really work. I will describe a new technique for analyzing these schemes---maintaining analytic control over large, noisy quantum systems---that leads to a rigorous proof that 0.1% gate depolarizing noise is tolerable (in a nonlocal gate model), lending support to the simulations. If the noise model is known, then the rigorous bound is as high as 1%. BIO: Ben Reichardt is a postdoctoral fellow at the Institute for Quantum Information, at the California Institute of Technology. He has a B.S. in mathematics from Stanford University, and a Ph.D. in computer science from the University of California, Berkeley. He studies quantum fault tolerance and quantum algorithms.http://www.its.caltech.edu/~breic/HOST: Todd Brun, tbrun@usc.edu
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited

Contact: Mayumi Thrasher
Apr
09

Application of Omics-Based Tools and Microarrays to Optimize Bioremediation
Wed, Apr 09, 2008 @ 02:00 PM - 03:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars

Lisa Alvarez-Cohen
Professor and Chair of Civil and Environmental Engineering Department, UC Berkeley This talk will focus on the application of omics-based molecular tools to understand and optimize the bioremediation of environmental contaminants by naturally occurring microorganisms. Bioremediation is a process that often relies upon mixed microbial communities to catalyze important biodegradation pathways. In order to improve our understanding of bioremediation processes, there is a need to develop rapid, quantitative, non-culture based molecular and microbiological assays for use in characterizing microbial communities and activities in subsurface soils. This talk will describe a variety of detection and quantification techniques applied to both genes and transcripts of interest, including quantitative PCR (qPCR), reverse transcription qPCR and whole-genome microarrays for comprehensive comparative genomics and transcriptomics with subsequent confirmation by proteomics. In addition, application of phylogenetic microarrays for identification of specific organisms in complex communities will be described. We have applied these tools to both laboratory samples for fundamental research, and to field samples for applied research in bioremediation of conventional and emerging contaminants.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited

Contact: Evangeline Reyes
Apr
09

NanoRobotic Systems
Wed, Apr 09, 2008 @ 03:30 PM - 04:30 PM
Aerospace and Mechanical Engineering
Conferences, Lectures, & Seminars

Lixin DongDr. Senior Research Scientist, Head of NanoRoboticsGroupInstitute of Robotics and Intelligent Systems (IRIS)Swiss Federal Institute of Technology (ETH), ZurichZurich, SwitzerlandAbstract:Progress in robotics over the past years has dramatically extended our ability to explore the world at a variety of scales extending from the edges of the solar system down to individual atoms. At the bottom of this scale, technology has been moving toward greater control of the structure of matter, suggesting the feasibility of achieving thorough control of the molecular structure of matter atom by atom. Nanorobotics represents the next stage in miniaturization for maneuvering nanoscale objects. Nanorobotics is the study of robotics at the nanometer scale, and includes robots that are nanoscale in size and large robots capable of manipulating objects that have dimensions in the nanoscale range with nanometer resolution. Nanorobotic systems emphasize the engineering aspect of nanorobotics and include the manufacturing and application technologies of nanorobotic manipulation systems, nanoelectromechanical systems (NEMS), and nanorobots (nano-sized robots, which have yet to be realized).
The well-defined geometry, exceptional mechanical properties, and extraordinary electrical characteristics of carbon nanotubes (CNTs) qualify them for structuring such systems. Relative displacements between the atomically smooth, nested shells in multiwalled carbon nanotubes (MWNTs) can be used as robust nanoscale motion enabling mechanisms for applications such as bearings, oscillators, shuttles, switches, memories, syringes, and actuators. The hollow structures of CNTs can serve as containers, conduits, pipettes, and coaxial cables for storing mass and charge, or for transport. On the other hand, novel helical nanostructures are created through a top-down fabrication process in which a strained nanometer thick heteroepitaxial bilayer such as SiGe/Si and InGaAs/GaAs curls up to form 3D structures with nanoscale features such as tubes, coils, rings, and spirals. Because of their interesting morphology, mechanical, electrical, and electromagnetic properties, potential applications of these nanostructures include springs, electromechanical sensors, magnetic field detectors, chemical or biological sensors, and inductors.
Shrinking device size to nanometer scales presents many fascinating opportunities such as manipulating nanoobjects with nanotools, measuring mass in zeptogram ranges, sensing forces at piconewton scales, and inducing gigahertz motion, among other new possibilities waiting to be discovered. Nanorobotic manipulation is a promising technology for structuring, characterizing and assembling nano building blocks into NEMS. Combined with recently developed nanofabrication processes, the technological progress on building nanorobotic systems from shell engineered CNTs and rolled up SiGe/Si and InGaAs/GaAs helical nanostructures is presented focusing on nanotube linear servo motors, nanorobotic spot welders using copper-filled nanotubes, and helical nanobelt motion converters.Lixin Dong is Senior Research Scientist at Swiss Federal Institute of Technology (ETH, Zurich), where he leads the NanoRobotics Group in the Institute of Robotics and Intelligent Systems (IRIS). He received the B.S. and M.S. degrees in Mechanical Engineering from Xi'an University of Technology (XUT) in 1989 and 1992, respectively. He became Research Associate in 1992, Lecturer in 1995, and Associate Professor in 1998 at XUT. He has served as the head of the Department of Mechatronics Engineering at XUT from 1997 to 1999. He received his Ph.D. degree in Micro Systems Engineering from Nagoya University in 2003, and became Assistant Professor at Nagoya University in 2003. In 2004 he joined ETH Zurich as a Research Scientist. His main research interests include nanorobotics, mechatronics, nanoelectromechnical systems (NEMS), mechnochemistry, and biomedical devices. He received the IEEE T-ASE Googol Best New Application Paper Award in 2007, Best Conference Paper Award at the Int. Conf. on Control Sci. and Engr. (ICCSE2003), and Finalist in the Best Paper Competition at IEEE-ICRA2007, IROS2005, and ICRA2001. He has been awarded the Science and Technology Advancement Prize by the Ministry of Education of China in 1999, by Shaanxi Province Government in 1999 and 1995, by Xi'an City Government in 1999, and by the Ministry of Machine-Building Industry of China in 1998 and 1992. He serves on the editorial board of the IEEE Trans. on Nanotechnology and the IEEE Trans. on Automation Science and Engineering.
Location: Seaver Science Library, Rm 150
Audiences: Everyone Is Invited

Contact: April Mundy
Apr
09

CS Colloq: Learning^3: Multi-Agent, Teacher-Agent, and Tutor-Student
Wed, Apr 09, 2008 @ 03:30 PM - 04:30 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars

Title: Learning^3: Multi-Agent, Teacher-Agent, and Tutor-StudentSpeaker: Dr. Yu-Han Chang (ISI)Abstract:
Learning is crucial aspect of any intelligent agent. The bulk of this talk with focus on our results in multi-agent learning, where agents must learn to adapt in environments populated with other adaptive, autonomous agents. I'll also spend some time briefly describing new projects in teachable agents, where agents can learn more rapidly by receiving interactive human instruction, and adaptive tutoring systems, where the tutoring system must learn to adapt to differing student capabilities and styles. In multi-agent environments, learning must account for the adaptive nature of the other agents. Traditional models such as MDPs, POMDPs, and game theoretic equilibria each have their shortcomings in this domain: e.g. the environment is not Markov, or the other agents may not be entirely rational. Regret is a principled framework for evaluating the performance of multi-agent learning algorithms, and regret-minimizing algorithms offer a good approach to this domain, one that does not need to make strong assumptions regarding expected types of opponents. I'll describe an algorithm that exhibits good performance against a wide range of possible opponents, and guarantees low regret against any arbitrary opponent.Biography:
Dr. Yu-Han Chang is a Computer Scientist at the Information Sciences Institute of the University of Southern California. His current research interests range from reinforcement learning and game theory to natural language understanding and interactive games. Recent and ongoing projects include using machine learning to improve education, "learning by noticing", planning in continuous battle spaces, training intelligent agents via interactive games, and developing no-regret algorithms for learning in non-cooperative domains. Dr. Chang holds undergraduate degrees in Mathematics and Economics, as well as a S.M. in Computer Science, from Harvard University. He received his Ph.D. in Electrical Engineering and Computer Science from MIT, where he developed algorithms for multi-agent learning in the context of machine learning and game theory.
Location: Henry Salvatori Computer Science Center (SAL) - 322
Audiences: Everyone Is Invited

Contact: CS Colloquia
Apr
10

Development of a biomimetic lung surfactant
Thu, Apr 10, 2008 @ 12:45 PM
Mork Family Department of Chemical Engineering and Materials Science
Conferences, Lectures, & Seminars

Lyman Handy Colloquium SeriesPresentsAnnelise E. BarronAssociate Professor, Dept. of Bioengineering, Stanford University, Stanford, CA 94305AbstractWe are developing a new family of amphipathic peptide mimics for a synthetic lung surfactant (LS) replacement. Presently used exogenous LS replacements are extracted from animal lungs and used to treat respiratory distress sydrome in premature infants. The hydrophobic lung surfactant proteins SP-B and SP-C are necessary constituents of an effective surfactant replacement for the treatment of respiratory distress. As there are concerns and limitations associated with animal-derived surfactants, much recent work has focused on synthetic peptide analogues of SP-B and SP-C. However, creating an accurate peptide mimic of SP-C that retains good biophysical surface activity is challenging, given this lipopeptide's extreme hydrophobicity and propensity to misfold and aggregate. One approach that overcomes these difficulties is the use of helical poly-/N/-substituted glycines, or "peptoids," to mimic SP-C. We discuss advances in the design and characterization of peptoid-based SP-C mimics, which recently have led to the creation of our most biomimetic surfactant replacements to date.
Peptoid sequences were systemically varied in order to study surface activity effects of varying peptoid helicity,/ N/-terminal side chain chemistry and sequence length, as well as the side chain structures used within the hydrophobic C-terminal helix. The secondary structures of the peptoid SP-C mimics are analyzed in organic solution by CD spectroscopy. Langmuir-Wilhelmy surface balance experiments, epifluorescence videomicroscopy studies, and pulsating bubble surfactometry are used to characterize the surface activity and surface film morphology of the mimics in combination with a biomimetic phospholipid formulation. These results provide us with the first comprehensive structure-function relationships for peptoid-based analogues of surfactant protein C, as well as strong evidence that they offer significant promise for use in a synthetic replacement for animal-derived surfactants. There are several other potential applications for a safe and non-immunogenic surfactant formulation with these properties, other than treating respiratory distress, including protection against ventilator-induced lung injury, drug delivery to the lungs, and treatment of ear infections.
Location: Olin Hall of Engineering (OHE) - 122
Audiences: Everyone Is Invited

Contact: Petra Pearce Sapir
Apr
10

CS Colloq: Dynamics of Real-World Networks
Thu, Apr 10, 2008 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars

Title: Dynamics of Real-World NetworksSpeaker: Jure Leskovec (CMU)Abstract:
Emergence of the web and cyberspace gave rise to detailed traces of human social activity. This offers great opportunities to analyze and model behaviors of millions of people. For example, we examined ''planetary scale'' dynamics of a full Microsoft Instant Messenger network that contains 240 million people, with more than 255 billion exchanged messages per month (4.5TB of data), which makes it the largest social network analyzed to date. In this talk I will focus on two aspects of the dynamics of large real- world networks: (a) dynamics of information diffusion and cascading behavior in networks, and (b) dynamics of the structure of time evolving networks. First, I will consider network cascades that are created by the diffusion process where behavior cascades from node to node like an epidemic. We study two related scenarios: information diffusion among blogs, and a viral marketing setting of 16 million product recommendations among 4 million people. Motivated by our empirical observations we develop algorithms for detecting disease outbreaks and finding influential bloggers that create large cascades. We exploit the ''submodularity'' principle to develop an efficient algorithm that finds near optimal solutions, while scaling to large problems and being 700 times faster than a simple greedy solution. Second, in our recent work we found counter intuitive patterns that change some of the basic assumptions about fundamental structural properties of networks varying over time. Leveraging our observations we developed a Kronecker graph generator model that explains processes governing network evolution. Moreover, we can fit the model to large networks, and then use it to generate realistic graphs and give formal statements about their properties. Estimating the model naively takes O(N!N^2) while we develop a linear time O(E) algorithm.Biography:
Jure Leskovec (www.cs.cmu.edu/~jure) is a PhD candidate in Machine Learning Departmen at Carnegie Mellon University. He is also a Microsoft Research Graduate Fellow. He received the ACM KDD 2005 and ACM KDD 2007 best paper awards, won the ACM KDD cup in 2003 and topped the Battle of the Sensor Networks 2007 competition. Jure holds three patents. His research interests include applied machine learning and large-scale data mining focusing on the analysis and modeling of large real-world networks as the study of phenomena across the social, technological, and natural worlds.
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited

Contact: CS Colloquia
Apr
11

Coding and Message-Passing for Large-Scale Distributed Storage and Inference
Fri, Apr 11, 2008 @ 10:30 AM - 11:30 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

SPEAKER: Alexandros G. Dimakis, EECS Dept, UC BerkeleyABSTRACT: Multiple recent advances in technology have catalyzed a paradigm shift away from centralized schemes and in the direction of distributed and cooperative architectures for large-scale systems. In applications like data centers, sensor networks, and peer-to-peer networks, coding is used to introduce redundancy for robustness. I will show that network coding can surprisingly reduce the communication requirements compared to standard Reed-Solomon codes used in current architectures. Further, I will present novel information theoretic performance bounds and explicit network codes that achieve optimal performance.For the case of large-scale distributed inference, I will present some novel message-passing algorithms and show explicit results on convergence rate. In particular, I will present the first gossip algorithm that scales linearly in the number of nodes for a large class of geometric graphs, resolving an open problem in this active new research area.BIO: Alex Dimakis is a Ph.D. candidate in the department of Electrical Engineering and Computer Sciences at UC Berkeley working with Prof. Martin Wainwright and Prof. Kannan Ramchandran. He received the Diploma degree in Electrical and Computer Engineering from the National Technical University of Athens in 2003 and the M.S. degree in Electrical Engineering from UC Berkeley in 2005. His research interests include Communications, Signal Processing, and Networking with applications in distributed systems and sensor networks. Mr. Dimakis has received two outstanding paper awards, the UC Berkeley Departmental Fellowship in 2003, and the Microsoft Research Fellowship in 2007.HOST: Prof. Keith Chugg, chugg@usc.edu
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited

Contact: Mayumi Thrasher
Apr
11

Honors Colloquium Lecture
Fri, Apr 11, 2008 @ 01:00 PM - 01:50 PM
Viterbi School of Engineering Student Affairs
Conferences, Lectures, & Seminars

Lecture offered by Mr. David M. Bowman, Vice President & C-17 Program Manager for The Boeing Company, Global Mobility Systems
Location: Olin Hall of Engineering (OHE) - 122
Audiences: Honors Program Students and all Faculty and Staff are invited to attend

Contact: Erika Chua
Apr
11

Low-Cost High-Efficiency Distributed Hydrogen Production for Combined Heat and Power
Fri, Apr 11, 2008 @ 01:00 PM - 02:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars

Dr. Durai Swamy, Intelligent Energy, Inc.Intelligent Energy is developing a new-generation 1-10 kWe CHP unit that achieves the high efficiency, high durability and low cost targets simultaneously includes a bold optimization and integration of existing IE technology platforms. The CHP unit will be based on IE's open architecture integration philosophy that maintains a high purity hydrogen interface between the hydrogen generation and fuel cell subsystems. The fuel cell subsystem will be derived from IE's 2kWe CHP platform and its advanced 10kWe auxiliary power unit platform that achieves 60% efficiency on pure hydrogen. An innovative hydrogen generation subsystem will be developed to support the aggressive costs and performance targets, but will leverage IE's experience from two validated technology platforms: - IE's 100 to 500We membrane reformer that achieves 99.9+% purity on seven different fuel types, and its 10kWe steam reformer integrated with a fast cycle pressure swing adsorption hydrogen purification system. Both of these H2 generation platforms achieve only efficiencies in the range of 60-65% currently. IE plans to investigate significant improvements in these technologies to increase the H2 generation efficiency over 75%.
The greatest challenge of the development will be to achieve an optimized balance between increased stack performance (high cell voltage at high current densities), low cost cell components, increased hydrogen generation efficiency (high fuel conversion, lower steam/carbon ratios, maximum recuperation of heat and water vapor, and high hydrogen recovery factors), low parasitic power components and efficient grid connected inverter, and least cost balance of plant in a fully integrated system.

Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited

Contact: Evangeline Reyes
Apr
11

Opportunistic Scheduling with Reliability Guarantees in Cognitive Radio Networks
Fri, Apr 11, 2008 @ 03:00 PM - 04:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

Speaker: Rahul UrgaonkarAbstract: We consider a cognitive radio network with static primary users (that are the licensed owners of the spectrum) and potentially mobile secondary users that try to send their data to the access points by utilizing idle primary channels. We develop opportunistic transmission scheduling policies for such networks that maximize the throughput utility of the secondary users subject to maximum collision constraints with the primary users. We use the technique of Lyapunov Optimization to design an online flow control, scheduling and resource allocation algorithm that meets the desired objectives and provides explicit performance guarantees. Specifically, our algorithm provides tight reliability guarantees in the form of a bound on the worst case number of collisions suffered by a primary user in any time interval. This algorithm operates without requiring a-priori knowledge of the mobility patterns of the secondary users and yields an average throughput utility that can be pushed arbitrarily close to the optimal value, with a trade-off in average delay.Bio: Rahul Urgaonkar obtained the B.Tech. degree in Electrical Engineering from the Indian Institute of Technology (IIT) Bombay in 2002 and the M.S. degree in Electrical Engineering from the University of Southern California, Los Angeles in 2005. He is currently a PhD student in Electrical Engineering at USC. His research interests are in the areas of stochastic network optimization, resource allocation, and scheduling in next generation Wireless Networks.

Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited

Contact: Gerrielyn Ramos
Apr
14

Martin Gunderson, Professor of Electrical Engineering, Physics, and Astronomy
Mon, Apr 14, 2008 @ 12:30 PM - 01:50 PM
Alfred E. Mann Department of Biomedical Engineering
Conferences, Lectures, & Seminars

Studies of nanosecond pulsed electric fields for the induction of apoptosis in cancer cells in vitro and tumors in vivo
Audiences: Department Only

Contact: Mischalgrace Diasanta
Apr
14

Capture and Reproduction of Spatial Audio
Mon, Apr 14, 2008 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

Speaker: Ramani Duraiswami, Perceptual Interfaces and Reality Lab, University of Maryland.Abstract: Our ability to perceive space using sound has been understood for a while, but not much exploited in audio systems designed for entertainment, augmented/virtual reality or data display. In this talk I will focus on our research in various aspects of spatial audio and scene understanding. Spherical microphone arrays provide a means of spatial audio capture and sound scene analysis. I will discuss several spherical arrays we have built, their theoretical analysis, and their use in beamforming sound, and in representing spatial sound. Of particular interest is the combination of arrays and video cameras, to obtain more complete scene understanding. We show that the spherical array can be treated as a camera, and combined with normal video cameras to provide geometric scene understanding. I will also discuss the recreation of spatial auditory reality over headphones, using the captured sound and head related transfer functions. A novel method for measurement of HRTFs invented in our lab will also be presented. Bio: Ramani Duraiswami, is an Associate Professor in the Department of Computer Science and at the Institute for Advanced Computer Studies at the University of Maryland. He has affiliate appointments in the Departments of Electrical Engineering, and in the Applied Mathematics and Scientific Computing Program. Prof. Duraiswami is director of the Perceptual Interfaces and Reality Lab. and has broad research interests in computational audition, scientific computing, computer vision and machine learning and has over 130 publications in these areas. He received an undergraduate degree from the Indian Institute of Technology, Bombay; and a Ph.D. from The Johns Hopkins University. More information on his research can be found at www.umiacs.umd.edu/~ramani Host: Prof. Shrikanth Narayanan
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited

Contact: Mary Francis
Apr
15

CS Colloq: Learning Low Dimensional Representations of High Dimensional Data
Tue, Apr 15, 2008 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars

Title: Learning Low Dimensional Representations of High Dimensional DataSpeaker: Dr. Fei Sha(UC Berkeley)Abstract:
Statistical modeling of high-dimensional and complex data is a challenging task in machine learning. To tackle this problem, a very powerful strategy is to identify and exploit low-dimensional structures intrinsic to the data. For example, text and image data can often be represented as suppositions of meaningful and interpretable structures such as ``object parts'' and ``topics''. These structures are composed of visually salient image patches as well as groups of semantically related words. Examples of such learning algorithms include nonnegative matrix factorization (NMF) and latent Dirichlet allocation (LDA), where parts and topics are encoded by nonnegative basis matrices and probability distributions respectively. In this talk, I will focus on my research that have brought new and interesting developments into the frameworks of NMF and LDA. In the first project, I show how to extend the original NMF approach to learning meaningful ``audio parts'' from speech and audio data. The audio parts robustly encode harmonic structures in the voices, which are key acoustic features for building machines that can analyze complicated acoustic signals as well as human listeners. In the second project, I investigate how to incorporate supervisory information like class labels in LDA models. In the supervised LDA, topics are discovered by grouping words based on not only semantic similarity but also class label proximity. These topics yield compact representation with better predictive powers than those derived from the original unsupervised LDA. Towards the end of the talk, I will summarize briefly my work on learning other types of latent structures such as manifolds and clusters. I will then conclude by discussing all these approaches in a general perspective and speculating a few interesting directions for future work.
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited

Contact: CS Colloquia
Apr
16

Multiscale computations for flow and transport in heterogeneous porous media
Wed, Apr 16, 2008 @ 02:00 PM - 03:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars

Speaker: Dr. Tom Hou, Charles Lee Powell Professor, Caltech Abstract: Many problems of fundamental and practical importance contain multiple scale solutions. Composite materials, flow and transport in heterogeneous porous media, and turbulent flow are examples of this type.
Direct numerical simulations of these multiscale problems are extremely difficult due to the wide range of length scales in the underlying physical problems. In this talk, I describe some of our recent efforts in developing multiscale computational methods to upscale two-phase flows in strongly heterogeneous porous media. For some challenging problems with long range scale interaction arising from engineering applications, we show how to use limited global information to improve the accuracy of the multiscale method. Another important application is how to quantify uncertainty in modeling the heterogeneous random media. We show that by using a coarse multiscale model to precondition the Markov Chain Monte Carlo method, we can significantly improve the efficiency of the MCMC method in generating the probability distribution of the random media subject to some production data. Finally, we introduce a new multiscale analysis for convection dominated incompressible flow with multiscale solutions. Our multiscale analysis provides an important guideline in designing a multiscale method for computing incompressible flow with multiscale solutions.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited

Contact: Evangeline Reyes
Apr
16

Novel Nanoscale Materials for the National Ignition Facility
Wed, Apr 16, 2008 @ 03:30 PM - 04:30 PM
Aerospace and Mechanical Engineering
Conferences, Lectures, & Seminars

Juergen Biener Nanoscale Synthesis and Characterization LaboratoryLawrence Livermore National LaboratoryLivermore, California 94550 Current designs of targets for the National Ignition Facility (NIF) require the development of novel nanoscale materials such as ultra-low density nanoporous metal foams, nanocrystalline metal and diamond films. This talk will provide an overview of the challenges and opportunities associated with the synthesis of these nanoscale materials, and discuss applications beyond NIF targets.
Prepared by LLNL under Contract DE-AC52-07NA27344

Location: Seaver Science Library, Rm 150
Audiences: Everyone Is Invited

Contact: April Mundy
Apr
17

Physics at the Nanoscale: Tubes, Sheets, Ribbons, and Junctions
Thu, Apr 17, 2008 @ 12:45 PM
Mork Family Department of Chemical Engineering and Materials Science
Conferences, Lectures, & Seminars

Mork Family Department of Chemical Engineering and Materials Science Distinguished Lecture SeriesPresentsProfessor Steven G. LouieDepartment of Physics
Materials Sciences Division
Lawrence Berkeley National Laboratory
University of California at Berkeley
Berkeley, CaAbstract The restricted geometry of nanostructures often gives rise to novel, unexpected properties and phenomena. In particular, symmetry and many-electron effects can become significantly more important in determining the behaviors of these systems. In this talk, I discuss some recent progress on using theory and computation to understand and predict some of their electronic, transport, optical, and mechanical properties. Examples of systems of interest include carbon and BN nanotubes, graphene, graphene nanoribbons, and molecular junctions. These nanostructures exhibit a number of unexpected behaviors – novel conductance characteristics, extraordinarily large excitonic effects (even in the metallic systems), interesting friction forces, anomalous anisotropy in the dynamics of carriers (the 2D massless Dirac fermions) in graphene under an external periodic potential, and an electric field-induced half-metallic state for the zigzag graphene nanoribbons, among others. The physical mechanisms behind these unusual behaviors are examined.Thursday, April 17, 2008
Seminar at 12:45 p.m.
OHE 122Refreshments served after the seminar in HED Lobby
The Scientific Community is Cordially Invited.
Location: Olin Hall of Engineering (OHE) - 122
Audiences: Everyone Is Invited

Contact: Petra Pearce Sapir
Apr
18

Quantum Information Processing: Spingineering the Future
Fri, Apr 18, 2008 @ 10:30 AM - 11:30 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

SPEAKER: Andrew J. Landahl, Research Assistant Professor, Information Physics Group, University of New MexicoABSTRACT: The spin of a quantum particle is a natural candidate for storing quantum information. A number of technologies are based on this idea, including quantum dots, optical lattices, and nuclear magnetic resonance. Devising new protocols for acquiring, processing, and transmitting information in these systems is a "spingineering" task that current and future generations of information technology researchers will face.I will present my work on three specific "spingineering" research problems, one each from the areas of communication, computation, and error correction. First I will show how to engineer a spin network to serve as a "quantum communication bus" that, in the absence of noise, allows arbitrary-distance perfect-fidelity quantum communication. Then I will show how to engineer a spin chain to serve as a "programmable universal quantum computer" that can execute an arbitrary program on an arbitrary input, where both the program and the data are encoded in the initial state of the spins. Finally, I will show how to perform "continuous-time quantum error correction" on a spin network by optimizing a feedback loop that uses weak measurements and Hamiltonian controls.These problems demonstrate that "spingineering" requires a broad-based theoretical background to achieve an engineering solution; in this talk the solutions will rely on quantum mechanics, algebraic coding theory, random walk theory, and feedback control theory.BIO: Andrew Landahl is a Research Assistant Professor in the Information Physics Group at the University of New Mexico. He was a Hewlett-Packard/MIT Postdoctoral Fellow from 2002-05 under the supervision of Professor Seth Lloyd. He earned Physics M.S. and Ph.D. degrees at Caltech (2000 and 2002) under the supervision of Professor John Preskill and earned Physics and Mathematics B.S. degrees, Summa cum Laude, at Virginia Tech (1996), with an Honors Thesis supervised by Professor Lay-Nam Chang.Professor Landahl's research spans the areas of quantum computing, quantum information, and quantum control. His research accomplishments include the fastest quantum algorithm for searching an ordered list, a spin-chain bus enabling lossless quantum communication, a topologically-protected fault-tolerant quantum memory, a printed quantum circuit architecture, and the theory of continuous-time quantum error correction. His refereed publications average over 40 citations per paper. He has supervised or co-supervised the research of seven students.Professor Landahl is a member of the American Physical Society (APS), Institute of Electrical and Electronics Engineers (IEEE), and the Association for Computing Machinery (ACM). He served as a Grand Awards Judge for the 2007 Intel International Science and Engineering Fair (ISEF) and mentored a Siemens-Westinghouse Regional Finalist in 2003-04. He currently serves as Chair of the Local Organizing Committee for QIP 2009, a high-profile annual international conference of over 350 researchers in the field of quantum information to be held in Santa Fe, NM in January 2009.HOST: Prof. Todd Brun, tbrun@usc.edu
Location: Frank R. Seaver Science Center (SSC) - 319
Audiences: Everyone Is Invited

Contact: Mayumi Thrasher
Apr
18

Stimulating the indigenous microorganisms to remove heavy metals from marine sediments
Fri, Apr 18, 2008 @ 01:00 PM - 02:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars

Jinjun KanDepartment of Earth Sciences
University of Southern California
Los Angeles CA 90089Microbial activities play critical roles in metal remediation in natural environments. Adding organic or inorganic amendments will stimulate growth of indigenous microorganisms and then the microbes contribute to sorb, degrade, transform or immobilize metals from the environments. In this current study, we evaluated microbial stimulation of adding inorganic (e.g. apatite) and organic (e.g. chitin, acetate) amendments to marine sediments. Significant bacterial biomass and activities were induced by amendments of apatite, chitin and acetate. Molecular fingerprints of bacterial communities by denaturant gradient gel electrophoresis (DGGE) showed that distinct bacterial populations occurred in different amendments, and the stimulated microorganisms contained three major bacterial groups, Alphaproteobacteria (Roseobacter), Gammaproteobacteria and Bacteroidetes. Alphaproteobacteria (Roseobacter) dominated in water columns while Bacteroidetes were predominant population in the sediments. Most probable number (MPN) analyses showed that sulfate-reducing bacteria were significantly induced in the treatments with chitin, apatite+chitin. Testing N-acetylglucosamine and acetate as potential carbon source has also recovered sulfate-reducing bacteria. Sulfide-producing and metal-reducing bacteria were also recovered by using thiosulfate as an electron-acceptor. The results indicate that amendments stimulate geochemically important bacteria and further investigations on their physiological properties are critical to recognize their actual roles in the metal remediation process.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited

Contact: Evangeline Reyes
Apr
18

CS Colloq: Efficient, Adaptive Inference for Distributions on Permutations
Fri, Apr 18, 2008 @ 04:00 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars

Title: Efficient, Adaptive Inference for Distributions on PermutationsSpeaker: Prof. Carlos Guestrin (CMU)Abstract:
Permutations are ubiquitous in many real world problems, such as voting, rankings and data association. Representing uncertainty over permutations is challenging, since there are $n!$ possibilities, and typical compact representations, such as graphical models, cannot efficiently capture the mutual exclusivity constraints associated with permutations. In this talk, we use the ``low-frequency'' terms of a Fourier decomposition to represent such distributions compactly. We first describe how the two standard probabilistic inference operations, conditioning and marginalization, can be performed entirely in the Fourier domain in terms of these low frequency components, without ever enumeration $n!$ terms. We also describe a novel approach for adaptively picking the complexity of this representation in order control the resulting approximation error. We demonstrate the effectiveness of our approach on a real camera- based multi-people tracking setting.Biography:
I am an assistant professor in the Machine Learning Department and in the Computer Science Department at Carnegie Mellon University. I co- direct the Sense, Learn, and Act (Select) Lab with Geoff Gordon. In 2003-2004, I spent a year as a senior researcher at the Intel Research Lab in Berkeley. In August 2003, I received my Ph.D. in Computer Science from Stanford University, where I was advised by Daphne Koller, in the DAGS research group. I received a Mechatronics Engineer (Mechanical Engineering, with emphasis in Automation and Systems) degree in 1998 from the Polytechnic School of the University of São Paulo, Brazil.
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited

Contact: CS Colloquia
Apr
21

Noah Malmstadt, Assistant Professor of Chemical Engineering and Materials Science, USC
Mon, Apr 21, 2008 @ 12:30 PM - 01:50 PM
Alfred E. Mann Department of Biomedical Engineering
Conferences, Lectures, & Seminars

Nanoscale structure formation in biomimetic cell membranes
Audiences: Department Only

Contact: Mischalgrace Diasanta
Apr
21

BME Weekly Seminar Series-Rehabilitation for the Self-Management of Parkinsons Disease
Mon, Apr 21, 2008 @ 04:00 PM - 05:00 PM
Alfred E. Mann Department of Biomedical Engineering
Conferences, Lectures, & Seminars

OBJECTIVE: To determine whether rehabilitation that focuses on self-management of health helps to improve the day-to-day functioning and quality of life of community-living clients with Parkinson's disease (PD), beyond the effects of medical treatment alone. DESIGN: In a randomized controlled design, 120 participants were assigned to one of three conditions for a duration of 6 weeks: (I) medication only, (II) medication plus 2 outpatient group rehabilitation sessions and 1 social activity session per week, or (III) medication plus 2 outpatient group rehabilitation sessions and 1 home/community rehabilitation session per week. Rehabilitation occurred through integrated physical, occupational, and speech therapy services specialized to the self-management of health needs of people with PD. RESULTS: Compliance with treatment was 93% in the outpatient setting and 99% in the home setting. The first specific aim of the study was to determine if increasing "doses" of self-management rehabilitation (from Conditions I to II to III) resulted in increasingly positive quality of life outcomes. The results support this dose effect (F(1,111) = 5.96, p = .005). With increasing doses of rehabilitation, individuals experienced increasingly positive quality of life outcomes as measured by the Parkinson's Disease Questionnaire- 39. DISCUSSION: These results are consistent with the findings from the meta-analyses and the previous randomized controlled trial conducted by our research group. The primary hypothesis of this study was supported: there was a linear increase in immediate post-intervention improvement in overall quality of life from optimal medication (Condition I) to the addition of18 hours of self-management rehabilitation (Condition II) to 27 hours of self-management rehabilitation (Condition III) administered over a six-week period. These effects were strongest immediately following intervention and persisted to a smaller degree at 6 months follow-up. For more information, please visit http://www.bu.edu/sargent/centers/research/neurorehab/team/index.shtmlThe seminar is simultaneously presented at UPS and HSC, and available via live Web Cast at:http://capture.usc.edu/college/Catalog/?cid=af180d48-ceff-42b9-a35c-eb199daed320Information about all seminars can be found at: http://bme.usc.edu/valero/seminar.htm
Location: HSC: B16 Basement of the Norris Medical Library ; UPC: Ahmanson Center for Biological Research (ACB)
Audiences: Faculty and Graduate Students

Contact: Toyicha Chisom
Apr
22

CS Colloq: Simple Encryption Schemes Against Sophisticated Attacks
Tue, Apr 22, 2008 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars

Title: Simple Encryption Schemes Against Sophisticated AttacksSpeaker: Dr. Hoeteck Wee (Columbia)Abstract:
Along with the increasing reliance on computers and the Internet for myriad tasks from voting to auctions evolves a pressing need to develop cryptographic tools and protocols with stronger guarantees. Traditional cryptographic guarantees such as data privacy amidst wiretapping and security against a static collection of malicious network entities do not meet the security requirements for many of these tasks: -- An adversary may be unable to learn your bid in an online auction if the bid is encrypted; however, it could potentially modify the ciphertext to obtain one corresponding to a bid that is a dollar higher than yours. -- An adversary that adaptively determines which electronic voting machines to break into during the course of an election has a better chance at influencing the outcome of an election than one that makes its choices before the election commences. I will present new constructions of encryption schemes addressing each of these attacks. The first scheme guarantees that given an encryption of a message, it is infeasible to generate an encryption of a related message. The second improves upon an important building block used in constructing protocols for general multi-party computation that are secure against an adversary that adaptively corrupts up to one third of the parties. Compared to most previous constructions, our schemes are simpler, more efficient, and can be realized under a larger class of cryptographic assumptions.Biography:
Hoeteck Wee is a postdoctoral researcher at Columbia University. He completed his undergraduate studies at MIT and his PhD at UC Berkeley under the supervision of Luca Trevisan. He was a visiting student at Tsinghua University (Beijing) from Aug 2005 till Jun 2006, a core participant in the program on Securing Cyberspace at IPAM (UCLA) in Fall 2006, and a (one-time photographic) contributor to the Schmap Amsterdam Guide.
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited

Contact: CS Colloquia
Apr
23

Assessing Accuracy and Significance of Structural and Functional Brain Images
Wed, Apr 23, 2008 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

SPEAKER: Dimitrios Pantazis, Postdoctoral Research Associate, Signal and Image Processing InstituteABSTRACT: Recent developments in instrumentation, coupled with new data analysis tools, have led to unique ways of noninvasively exploring the human brain. For example, whole head Magnetoencephalography (MEG) arrays provide dynamic images of human brain function at a millisecond scale; Magnetic Resonance Imaging (MRI) methods allow access to structural and diffusion brain images, as well as functional maps of hemodynamic response. I will discuss ways of identifying experimental effects in dynamic images of MEG brain activation through the use of random field theory and permutation statistical tests. I will then extend these tests to spatial-spectral activation maps, morphological tests of brain structures, and tumor detection in positron emission tomography. Efficient methods to analyze MEG data through the use of custom analysis of covariance models will be presented and applied in a human visual attention study. I will conclude with an evaluation of manual and automatic human cortex registration methods.BIO: Dimitrios Pantazis is Postdoctoral Research Associate in the Department of Electrical Engineering at the University of Southern California, and the Dornsife Cognitive Neuroscience Imaging Center and House Ear Institute in Los Angeles through the Biomedical Imaging Sciences Initiative in USC. He is a member of the Biomedical Imaging Research Lab lead by Prof. Richard M. Leahy, and has broad research interests in modeling and statistical analysis of anatomical and functional brain signals. He received a diploma in EE from the Aristotle University of Thessaloniki - Greece in 1996, and a M.S and Ph.D. in EE from the University of Southern California in 2003 and 2006 respectively. More information on his research can be found at http://neuroimage.usc.edu/pantazisHOST: Richard Leahy, leahy@sipi.usc.edu
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited

Contact: Mayumi Thrasher
Apr
23

Desensitizing Halfband Interpolation Filters
Wed, Apr 23, 2008 @ 11:00 AM - 12:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

Speaker: Professor Alan WillsonElectrical Engineering Department
University of California, Los Angeles, CaliforniaABSTRACTA very common component in digital circuitry for communications systems is the halfband filter. Halfband filters are often used in cooperation with up-samplers and down-samplers when a sampling-rate change is required. While techniques for designing these filters are well known, an entirely new design method has just been discovered wherein these filters can be made to possess a significant desensitivity to the filter's tap coefficient values. Such desensitivity can be exploited to yield halfband filters with reduced hardware requirements, which leads to circuits having lower power consumption, higher operating speeds, and smaller IC area. This talk will give a brief introduction to the concept of halfband filters and the applications of halfband filters. It will then explain the rationale and the method for the desensitizing of the filters and, finally, will illustrate through design examples and further explanation how the desensitivity improves upon the conventional designs.BIOGRAPHYDr. Alan N. Willson, Jr. is a Distinguished Professor of Electrical Engineering at UCLA. He served as Assistant Dean for Graduate Studies in the UCLA School of Engineering and Applied Science from 1977 through 1981, and served as Associate Dean of Engineering from 1987 through 2001. He has been engaged in research concerning computer-aided circuit analysis and design, the stability of distributed circuits, properties of nonlinear networks, theory of active circuits, digital signal processing, analog circuit fault diagnosis, and integrated circuits for signal processing. He is editor of Nonlinear Networks: Theory and Analysis (New York: IEEE Press, 1974). In 1991, he founded Pentomics, Inc. Dr. Willson served as editor of the IEEE Transactions on Circuits and Systems from 1977 to 1979 and as President of the IEEE Circuits and Systems (CAS) Society in 1984. He was the recipient of the 1978 and 1994 Guillemin-Cauer Awards of the IEEE CAS Society, the 1982 George Westinghouse Award of the American Society for Engineering Education, the 1982 Distinguished Faculty Award of the UCLA Engineering Alumni Association, the 1984 Myril B. Reed Best Paper Award of the Midwest Symposium on CAS, the 1985 and 1994 W. R. G. Baker Awards of the IEEE, the 2000 Technical Achievement Award and the 2003 Mac Van Valkenburg Award of the IEEE CAS Society. In 2007 he and his recent Ph.D. student, Guichang Zhong, were winners of the 44th DAC/ISSCC Student Design Contest.HOST: Professor Sanjit K. Mitra

Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited

Contact: Talyia Veal
Apr
23

Long Actuator Delays - Extending the Smith Predictor to Nonlinear
Wed, Apr 23, 2008 @ 03:30 PM - 04:30 PM
Aerospace and Mechanical Engineering
Conferences, Lectures, & Seminars

Miroslav KrsticHarold W. Sorenson Professor of Control SystemsDept. of Mechanical & Aerospace EngineeringUniversity of California at San DiegoAbstract: One would be hard pressed to find "long actuator delays" and "nonlinear control" co-existing in the same sentence in the existing control literature, which is due to the infinite dimensionality and the potential for finite escape time instability in the underlying problems. On the 50th anniversary of Otto Smith's invention of the "predictor" feedback for compensating long actuator delays for linear systems, a method that has since become one of the favorite tools in chemical process control and many other applications, I am pleased to present an approach for synthesizing a predictor feedback to go along with any stabilizing nominal nonlinear controller, with actuator delay of any length. Interestingly, Smith's idea was actually an elementary version of "infinite dimensional backstepping," which I have been developing over the last few years for PDE problems such as Navier-Stokes, MHD, Euler and Timoshenko beams, and other systems in mechanics. By employing the backstepping point of view to construct Lyapunov-Krasovskii functionals, it becomes possible to prove several forms of robustness of predictor feedbacks, including robustness to both underestimating and overestimating the length of the actuator delay. The latter is a particularly subtle result because it involves a non-standard dynamic perturbation - the controller (inadvertently) inserts an additional infinite-dimensional state to an already infinite-dimensional feedback loop.

Location: Seaver Science Library, Rm 150
Audiences: Everyone Is Invited

Contact: April Mundy
Apr
24

Green Campus and Education for Sustainability
Thu, Apr 24, 2008 @ 02:00 PM - 03:30 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars

Al Dorman Distinguished Lecture Series
Speaker: Dr. Jean-Lou Chameau President, Caltech Click link for further information:
http://viterbi.usc.edu/news/events/keynote/dorman/
Location: Ronald Tutor Hall of Engineering (RTH) - 526
Audiences: Everyone Is Invited

Contact: Evangeline Reyes
Apr
25

Studying Brain Dyanmics in Large-Scale Cortical Networks with MEG Imaging and Human Intracerebral...
Fri, Apr 25, 2008 @ 02:00 AM - 03:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars

Speaker: Karim JerbiNeurophysiology of Action and Perception Collège de France CNRS,Paris, France
&
Brain Dynamics and Cognition Lab, INSERM,
Lyon, France Abstract:Increasing evidence from human and non-human primate studies suggests that long-range cerebro-cerebral synchronization might reflect ongoing communication between distant neural populations essential for integrative behaviour. In the first part of this talk I will present the results of an MEG study in which we investigated the task-specific modulation of cerebral oscillations and long-distance cortico-cortical coupling in multiple frequency bands during a sustained visuomotor task. Our results show modulations of oscillatory power in various frequency bands in multiple areas including an increase in high gamma power (60-90 Hz) in motor and premotor areas during visuomotor control. Coherence analysis provides evidence for task-specific increases in low-frequency coherence between the primary motor cortex and multiple cortical and subcortical brain areas forming large-scale functional networks including the fronto-parietal circuit and the cerebello-thalamo-cortical pathway. In the second part of the talk I will provide an overview of some of the ongoing studies performed with intracerebral recordings in humans performing various cognitive tasks. Both online and offline analysis will be presented. The invasive experiments with real-time visual feedback of task-related brain power modulations (/Brain TV/ set-up) are discussed in the context of invasive brain-computer interfaces, neurofeedback as well as its utility basic neuroscience research.Host: Professor Richard Leahy
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited

Contact: Talyia Veal
Apr
25

Model, Identification & Analysis of Complex Stochastic Systems:
Fri, Apr 25, 2008 @ 11:00 AM - 01:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars

Sonjoy Das- PhD DefenseThe work presented in this dissertation focuses on characterization, identi¯-cation and analysis of stochastic systems. Stochastic systems refer to a physical
phenomena with inherent uncertainty in it, and are generally characterized by
two forms of representation: (1) a governing conservation law or partial dif-ferential equation (PDE) with some of its parameters interpreted as random
processes, and (2) a model-free random matrix operator. In this work, three
data-driven approaches are ¯rst introduced to characterize and construct consis-tent probability models of non-stationary and non-Gaussian random processes
or ¯elds within the polynomial chaos (PC) formalism. The resulting PC rep-resentations would be useful to probabilistically characterize the system input-output relationship for a variety of applications. Second, a novel hybrid physics-and data-based approach is proposed to characterize a complex stochastic sys-tems by using random matrix theory. Application of this approach to multiscale
mechanics problems is also presented. Also discussed in this work is a simple,
computationally e±cient and experiment-friendly coupling scheme based on fre-quency response function. This coupling scheme would be useful for analysis of
a complex stochastic system consisting of several subsystems characterized by,
e.g., stochastic PDEs or/and model-free random matrix operators.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited

Contact: Evangeline Reyes
Apr
25

Role of ammonia and water vapor on the formation of secondary organic aerosols in alkene/ozone react
Fri, Apr 25, 2008 @ 01:00 PM - 02:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars

Dr. Kwangsam NaBourns College of Engineering¡XCenter for Environmental Research and Technology (CE-CERT), University of California, RiversideAbstractThis study deals with the influence of ammonia (NH3) on secondary organic aerosol (SOA) formation from the ƒÑ-pinene/ozone systems in the dry and humid conditions. The effect of different OH scavengers on SOA formation is briefly described. The aerosol yield differed depending on which OH scavenger was used. Yield was the highest using CO, followed by cyclohexane and 2-butanol. It was found that the number and volume concentrations were quickly increased by 15% and 8%, respectively when NH3 was added after the reaction ceased. An increase in number concentration indicates the formation of new particles resulting from gas-to-particle conversion. Moreover, average size of particles increased from 242 nm to 248 nm. The resulting particles may be ammonium salts formed by the reaction between organic acids and NH3. When NH3 was added to an aerosolized cis-pinonic acid in the chamber reactor, a dramatic increase in both number and volume concentrations of cis-pinonic acid was observed. This is evidence that NH3 drives gas-phase organic acids into condensable salts, leading to elevated SOA formation. Initially present NH3 significantly enhanced aerosol yield in ƒÑ-pinene-ozone reactions, regardless of the presence of water vapor. However, the enhancing effect of NH3 on SOA formation was found to diminish in humid conditions. The degree of the increase in SOA yield in the presence of NH3 was higher in the humid condition than in the dry condition. The role of NH3 on SOA formation in the dry and humid conditions was further explained and discussed in terms of a theoretical modeling study.

Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited

Contact: Evangeline Reyes
Apr
28

Understanding Solute Transport in Extremely Heterogeneous Subsurface Media: A Field Perspective
Mon, Apr 28, 2008 @ 02:00 PM - 03:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars

Dr.Chunmiao Zheng- Department of Geological Sciences, University of AlabamaField studies at well-instrumented sites have played a preeminent role in our efforts to better understand and characterize solute transport processes in geologic media. In particular, several field tracer tests conducted at well-known sites such as those in Borden, Canada (MacKay et al., 1986), Mobile, Alabama (Molz et al., 1986), Twin Lake, Minnesota (Killey and Moltyaner, 1988), Cape Cod, Massachusetts (LeBlanc et al., 1991) and Columbus, Mississippi (Boggs et al., 1992) have provided new insights and extensive data sets for development and testing of transport theories and mathematical models. This presentation focuses on the Columbus, Mississippi site, more commonly known as the Macrodispersion Experiment (MADE) site.
Since the 1980s, three large-scale natural-gradient tracer tests and other types of field studies have been conducted at the MADE site located inside the Columbus Air Force Base in Mississippi, southeastern United States. The field data from the MADE site have been used extensively by researchers around the world to investigate and understand contaminant transport processes in highly heterogeneous aquifers. The MADE site has directly or indirectly inspired the development of a number of new and improved theories and computer models aimed at more accurate representation and prediction of contaminant transport affected by extreme aquifer heterogeneity. A great deal of recent and on-going research on contaminant transport in heterogeneous media has been motivated by the findings from the MADE site that suggest the existence of preferential flow paths resulting from small-scale heterogeneities. Since the 1980s, three large-scale natural-gradient tracer tests and other types of field studies have been conducted at the MADE site located inside the Columbus Air Force Base in Mississippi, southeastern United States. The field data from the MADE site have been used extensively by researchers around the world to investigate and understand contaminant transport processes in highly heterogeneous aquifers. The MADE site has directly or indirectly inspired the development of a number of new and improved theories and computer models aimed at more accurate representation and prediction of contaminant transport affected by extreme aquifer heterogeneity. A great deal of recent and on-going research on contaminant transport in heterogeneous media has been motivated by the findings from the MADE site that suggest the existence of preferential flow paths resulting from small-scale heterogeneities. This presentation will provide an overview of the research activities at the MADE site over the past 25 years and discuss how the findings from field studies have motivated various theories and models purported to accommodate so-called non-ideal transport behavior observed at the MADE site. Of particular interest are our recent field campaigns to further test the hypothesis that solute transport at the MADE site is dominated by small-scale preferential flow paths. These field campaigns include forced-gradient tracer tests, soil coring, direct-push hydraulic profiling, and high-resolution resistivity and GPR surveys. The on-going efforts are aimed at building stronger theoretical underpinnings and useful practical tools for modeling solute transport and for designing remedial measures in extremely heterogeneous aquifers such as that encountered at the MADE site.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited

Contact: Evangeline Reyes
Apr
29

CS Colloq: Random Projection Trees and Low Dimensional Manifolds
Tue, Apr 29, 2008 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars

Title: Random Projection Trees and Low Dimensional ManifoldsSpeaker: Prof. Sanjoy Dasgupta (UCSD)Abstract:
The curse of dimensionality has traditionally been the bane of nonparametric statistics (histograms, kernel density estimation, nearest neighbor search, and so on), as reflected in running times and convergence rates that are exponentially bad in the dimension. This problem is all the more pressing as data sets get increasingly high dimensional. Recently the field has been rejuvenated in several ways, of which perhaps the most promising is the realization that a lot of real-world data which appears high-dimensional in fact has low "intrinsic" dimension in the sense of lying close to a low-dimensional manifold. In the past few years, there has been a huge interest in learning such manifolds from data, and then using the learned structure to transform the data into a lower-dimensional space where standard statistical methods generically work better. I'll exhibit a way to benefit from intrinsic low dimensionality without having to go to the trouble of explicitly learning its fine structure. Specifically, I'll present a simple variant of the ubiquitous k-d tree (a spatial data structure widely used in machine learning and statistics) that is provably adaptive to low dimensional structure. We call this a "random projection tree" (RP tree). Along the way, I'll discuss different notions of intrinsic dimension -- motivated by manifolds, by local statistics, and by analysis on metric spaces -- and relate them. I'll then prove that RP trees require resources that depend only on these dimensions rather than the dimension of the space in which the data happens to be situated. This is work with Yoav Freund (UC San Diego).
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited

Contact: CS Colloquia
Apr
30

Environmental Engineering for Los Angeles County
Wed, Apr 30, 2008 @ 02:00 PM - 03:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars

Alex Mena & Mike Sullivan- Los Angeles County Sanitation DistrictsAbstract:
The Sanitation Districts of Los Angeles County (Districts) lead a regional movement towards environmental sustainability in California by converting waste to vital resources such as water and energy. The Districts have been providing wastewater and solid waste management since the late 1920's, and have been expanding renewable energy recovery options at every opportunity. The Districts currently provide these services for over five million people, and have an average annual budget that exceeds $1 Billion. Facilities include: over 1,300 miles of sanitary sewers; eleven wastewater treatment plants treating over 500 MGD; a reclaimed water reuse program serving over 500 sites directly, as well as replenishing groundwater; three active sanitary landfills handling a total of approximately 20,000 tons per day, including the largest landfill in the nation; three materials recovery facilities/transfer stations; and multiple energy production facilities, generating over 127 Megawatts. Energy recovery operations provide enough renewable energy to power 150,000 homes. The Districts are consistently nationally ranked in the top 3 of the "Top 10 Local Government" Green Power Partners selected by the EPA and are the 20th largest producer of electrical power in California.
If you have more space....Innovative projects in progress include design and construction of a regional 2,800 acre landfill that will accept 100 years of solid waste via rail transport, and the largest indoor and outdoor biosolids composting facilities in the country. The Districts are currently investigating the environmental benefits of constructing a new ocean outfall. If constructed, the $1-2 Billion outfall project would represent one of the most challenging and high profile public works projects in the nation. Engineers are involved in all phases of our work, from conception to operation. Almost all engineering is performed in-house, and the majority of Districts engineers have graduate degrees in civil/environmental engineering. Over 250 Districts engineers work in areas such as: research, air and water quality monitoring, design, geotechnical engineering, hydrogeology, planning, construction, operations, industrial waste, and financial management as well as the specialty areas of electrical, instrumentation, mechanical, and structural engineering.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited

Contact: Evangeline Reyes
Apr
30

Towards Dislocation Dynamics in Carbon Nanotubes and Graphene
Wed, Apr 30, 2008 @ 03:30 PM - 04:30 AM
Aerospace and Mechanical Engineering
Conferences, Lectures, & Seminars

This seminar, the last in the Spring AME Seminar Series, will be presented by Elif Ertekin from UC Berkeley.For more information including an abstract, go to http://ame-www.usc.edu/seminars/4-30-08-ertekin.shtml
Location: Seaver Science Library, Room 150 (SSL 150)
Audiences: Everyone Is Invited

Contact: Dennis Plocher

Events Calendar

Select a calendar:

Filter April Events by Event Type:

Conferences, Lectures, & SeminarsEvents for April

Conferences, Lectures, & Seminars
Events for April