Principal Investigators
University of New Mexico: Stephanie Forrest and Jared Saia
University of Virginia: John C. Knight (PI), Jack Davidson, David Evans,
Wes Wimer, Anh Nguyen-Tuong
University of California, Davis: Hao Chen, Karl Levitt, Jeff Rowe, Zhendong Su,
Felix Wu
University of California, Santa Barbara: F. Chong
Funding
Air Force Office of Scientific Research, Multidisciplinary Research
Program of the University Research Initiative (MURI)
Five years, $745,000 (UNM share)
Awarded July 1, 2007
Project Summary
Our continued and increasing reliance on information systems leaves society vulnerable to asymmetric cyber threats, in which a small group of determined adversaries can easily disrupt our cyber infrastructure. The project will develop a coherent body of scientific and engineering knowledge for constructing self-regenerative enterprise systems. Such systems have the following hallmarks: They are automated and proactive; They can adapt and reconfigure themselves to present attackers with an ever-shifting attack surface; And they can repair themselves when confronted with both known and unknown attacks.
Principal Investigators
Terran Lane
Kiri L. Wagstaff
Funding
National Science Foundation
Four Years,
$900,000
Project Summary
A new world of computing is being born, based on inexpensive, lightweight, powerful nodes that integrate computation, sensing, and communication. We envision a future in which many such nodes are embedded in the world around us, self-assembling into environmentally-aware networks that will assist us with workflows, education, entertainment, and safety. To achieve this vision, these ubiquitous computing networks will have to be able to integrate diverse streams of sensor information into coherent views of their environment. This project will initiate a new generation of machine learning methods to address the resulting data fusion and environmental awareness challenges. We will develop topology-aware machine learning methods that will (a) learn and exploit the topological structure of the environment and (b) enable collaborative learning among distributed learning agents. We will test these methods on live sensor networks currently installed at Kilauea and Mt. Erebus, two active volcanoes. We will involve students as investigators via middle-school student workshops at the Sally Ride Festival and via undergraduate and graduate student research assistantships. This work will transform both machine learning and ubiquitous computing by opening a vast space of novel machine learning problems and by inspiring the development of new capabilities for ubiquitous computing systems.
Principal Investigators
William McCune
Robert Veroff
Funding
National Science Foundation, Grant CNS-0708218
Three years, $400,000
Awarded July 24, 2007
Project Summary
The project is to design, develop, and disseminate software infrastructure to support research and education in automated deduction and formal methods. The core of the infrastructure will be software for high-performance first-order deduction. It will include libraries of modules for constructing special-purpose deduction systems as well as stand-alone programs that search for proofs and for counterexamples. The work will be easily accessible, and it will include documentation of the infrastructure in the form of application programmer interfaces, manuals, tutorials, and examples.
Principal Investigators
Fred Koster (Lovelace Respiratory Research Institute)
Stephanie Forrest (UNM)
Funding
National Institutes of Health 1 R21 AI 73607-01
Two years, $144,046 (UNM share)
Awarded April 1, 2007
Project Summary
The project will model early influenza virion productivity in human airway epithelial cells using both in vitro and computer models. The in vitro model uses primary human differentiated tracheobronchial and bronchiolo-alveolar epithelial cells grown in air-liquid interface (ALI) culture to document the kinetics of virion productivity of human pathogens, both H3N2 strains and H5N1 avian influenza strains. The agent-based computer model is a two-dimensional agent-based simulation. The in vitro and computer models are complementary, and the project will develop and use them synergistically. Virulence of influenza viruses is highly variable, but the causes of this variability are incompletely catalogued. Although molecular biology has provided a detailed understanding of the replication cycle in immortalized cells, influenza replication in intact tissue among phenotypically diverse epithelial cells of the human respiratory tract remains poorly understood, and this is the problem which the project will begin to address.
Principal Investigator
Joe Kniss
Funding
National Science Foundation, Grant CCF 0702787
Two Years, $200,000
Awarded July 2, 2007
Project Summary
This research develops visualization methods that aid decision-making by exposing important data characteristics and their inherent uncertainties in meaningful ways. Most visualization methods use some form of classification to eliminate unimportant regions and illuminate interesting ones. The process of classification is inherently uncertain; in general the source data contains noise, data transformations can further introduce and magnify uncertainty. This project will make the concept of uncertainty a central component of visualization. Users should be able to assess uncertainty, and therefore potential for error, while exploring the space of possibility with respect to concrete realizations of ambiguous features.
Principal Investigators
Shuang Luan (Investigator)
Cedric X. Yu (Department of Radiation Oncology, University of Maryland)
Funding
National Cancer Institute, R01CA117997
Awarded June 1, 2007
Total Amount: 4 Years, $260,000
Project Summary
Intensity-modulated radiation therapy (IMRT) is a state-of-the-art technique for modern radiation cancer treatment which aims to deliver a highly conformal radiation dose to a tumor while sparing the surrounding normal tissue and sensitive structures. Among the various IMRT techniques that have been proposed or implemented, the intensity-modulated arc therapy (IMAT) is one of the most promising approaches for achieving superior dose conformity with a minimized treatment time. However, due to the many degrees of freedom in IMAT planning, optimizing an IMAT plan is computationally difficult, and an effective method for IMAT planning is not yet available. Moreover, methods for IMAT (and other rotational delivery schemes) to handle breathing induced target tumor motion are also lacking. The goal of this project is to develop a clinically practical IMAT planning system that allows tumor tracking and image guidance through a bioengineering research partnership.
Principal Investigator
Jared Saia
Funding
National Science Foundation, CAREER Award CNS-0644058
Five years, $400,000
Awarded July 2007
Project Summary
This project will address fundamental problems in the area of robust collaborative computation. We will determine what can and can not be computed by a group of agents when an unknown subset of the agents are controlled by a computationally unbounded adversary. Solving these theoretical problems will likely have broader impact in such diverse areas as voting, spam detection, worm and malware detection, designing distributed file systems, auction and mechanism enforcement, collaborative filtering, and web search.
Principal Investigators
Jared Saia (UNM)
Daniel Rubenstein (Princeton Univ.)
Tanya Berger-Wolf (Univ. Illinois, Chicago)
Funding
National Science Foundation, Grant IIS-0705477
Three years, $300,000 (UNM share)
Awarded Aug. 6, 2007
Project Summary
Population biology is at the threshold of a new era. Recent breakthroughs in data collection technology offer the promise of answering some of the big questions about social animals. Which individuals are leaders and to what degree do these leaders control the behavior of others? How do social groups change when individuals join or leave the population or with external events such as the presence of a predator? To what degree can we predict future social interactions based on past information? The ultimate goals of this research proposal are to design a powerful and general abstract mathematical model that captures the distinct properties of dynamic social networks; to design rigorous, efficient, and scalable algorithms to answer queries within this model; and to validate the model and our algorithms using domestic and wild equid populations.
News & Events | People | Academics | Research | Jobs | Companies | Visitors | Computer Facilities | Home
