2005 - 2006 Program on National Defense and Homeland Security (NDHS)

Research Foci
Description of Activities
Working Groups
Further Information

For several years, groups of researchers have been seeking to define appropriate roles for the statistical sciences, applied mathematical sciences and decision sciences in problems of NDHS. Many efforts have focused on short-term applicability of existing methods and tools, rather than articulating or initiating a longer-term research agenda. Moreover, none of them has really spanned the statistical sciences, the applied mathematical sciences and the decision sciences. Perhaps most important point is that, despite progress, these efforts have not "jelled" to produce a self-sustaining research momentum in the statistical sciences, applied mathematical sciences and decision sciences on problems of NDHS. The SAMSI NDHS program is meant fill this gap, in part by providing proof of concept that the necessary collaborations are feasible.

Research in the SAMSI NDHS program will be cross-disciplinary, and many of the efforts it catalyzes will be addressable only by multi-institution teams of researchers. Some of the research needed to address problems in NDHS is technology-oriented – for example, development of sensors or biometric identification devices, but it is clear that data (statistical sciences), models (applied mathematics) and decisions (decision sciences) are essential components of the effort. These components interact in both obvious and subtle ways, some of which are just beginning to be understood, and which fit precisely the mission of SAMSI.

Program Leaders Committee: James Crowley (SIAM), Lawrence Cox (National Center for Health Statistics; co-chair), Alan F. Karr (NISS; Directorate Liaison), Sallie Keller-McNulty (Rice University; National Advisory Committee Liaison), Jon Kettenring (Drew University), and Nell Sedransk (NISS; co-chair)

Scientific Committee: Alicia Carriquiry (Iowa State University), Philip Hanlon (University of Michigan), Carey Priebe (Johns Hopkins University), Fred Roberts (DIMACS), Stephen M. Robinson (University of Wisconsin) and Ed Wegman (George Mason University), as well as others to be designated.

Research Foci

Although the ultimate foci of the program will be determined by those who participate, potential domain science and theory/methodology foci include:

• Biointelligence, which intersects planned development of a CDC Biointelligence Center. Such a center would need to analyze data from a variety of sources, which have differing characteristics in terms of temporal and spatial resolution, seasonal and regional variation, accuracy, completeness and complexity.
• Real Time Inference, also known as data streams. Clearly many of these problems present deep questions of estimation and control, and so naturally require collaboration of statistical and applied mathematical scientists.These are also decision problems, leading naturally engagement of the decision sciences and operations research.
• Anomaly detection, with particular attention to multivariate (possibly very high-dimensional) data, extremely rare events and false positives.
• Data integration, with attendant problems of privacy, confidentiality and "new forms" of data such as images or biometric identifications.
• Dynamics of massive databases, which is in part a fundamental issue of data quality. How long is it before an accumulating database (e.g., one containing facial images) becomes hopelessly contaminated. Is the contamination global or local? What strategies can retard or reverse the process?

Description of Activities

Workshops: The kickoff workshop, to be held September 11-15, 2005, will be similar in nature to other recent SAMSI kickoff workshops – highly participatory, with "Birds of a Feather" Sessions, a "Second Chance Seminar," a poster session, a young researchers session and many opportunities for individual discussions. Its principal goal will be to engage a broadly representative segment of the statistical, applied mathematical and decision sciences communities in formulation and pursuit of specific research activities to be undertaken by the NDHS Working Groups, which include:

• Formulation of central research issues
• Identification of testbed databases
• Formation of external partnerships between the Working Groups and others, especially Kickoff Workshop participants, interested in NDHS.

An upcoming transition workshop, date to be announced, will focus on:

• Presentation of results, a research agenda and problem lists generated by the NDHS program to the statistical sciences, applied mathematics and decision science communities;
• Formulation of follow-on activities for the program, and engage attendees who did not participate deeply in the program in these activities.

Education: While NDHS is not (yet, in any case) the subject of a "traditional" course, we plan a seminar course strongly focused on problems, problem owners and (potentially) testbed database, which will run either throughout the year or for one semester.

Outreach: NDHS is a natural and likely topic for SAMSI undergraduate workshops during 2005-06.

Working Groups

Further Information

• For research visits, including New Researcher Fellowships and SAMSI-University Fellowships, consult Visiting and Employment Opportunities for 2006-07.
• For postdoctoral appointments, consult Postdoctoral Positions for 2006-07.
• For other questions, send E-mail to [email protected].

Entire site © 2001-2008, Statistical and Applied Mathematical Sciences Institute. All Rights Reserved.