Workshop Details

The past thirty years have seen a huge research boom at the interface of statistical physics, combinatorics, probability, and the theory of computing. DIMACS has had a big role in this emergence. We'll take a look at how DIMACS helped get things started, what has been accomplished, and where we are going from here.

Speaker Bio: Peter Winkler is William Morrill Professor of Mathematics and Computer Science at Dartmouth College (but serving this academic year as Distinguished Visiting Professor of the Public Dissemination of Mathematics at the National Museum of Mathematics). Prior to joining Dartmouth, Winkler was Director of Fundamental Mathematics Research at Bell Labs. Working primarily in combinatorics, probability, and the theory of computing, he was one of the key figures in the extensive DIMACS activities in “statistical physics,” including the Special Focus on Discrete Probability and the Special Focus on Discrete Random Systems. He holds a dozen patents on topics that include cryptology, holography, optical networking, distributed computing, and marine navigation. Winkler received the David P. Robbins Prize of the Mathematical Association of America and delivered the Pi Mu Epsilon J. Sutherland Frame Lecture at MathFest in 2018. He is the author of two collections of mathematical puzzles, a portfolio of compositions for ragtime piano, a book on use of cryptography in the game of bridge, editor (with Jaroslav NešetÅ™il) of the DIMACS volume Graphs, Morphisms and Statistical Physics, and Chair of the DIMACS advisory Board.

Markov chain Monte Carlo methods have become ubiquitous across science and engineering as a means of exploring large configuration spaces. The idea is to walk among the configurations so that even though you explore a very small part of the space, samples will be drawn from a desirable distribution. Over the last 30 years there have been tremendous advances in the design and analysis of efficient sampling algorithms for this purpose, largely building on insights from statistical physics. One of the striking discoveries has been the realization that many natural Markov chains undergo a phase transition where they change from being efficient to inefficient as some parameter of the system is varied. In this talk we will explore this phenomenon, and show how insights from computing and probability reveal interesting results for asynchronous models of programmable matter.

Speaker Bio: Dana Randall is the ADVANCE Professor of Computing and co-Executive Director of the Institute for Data Engineering and Science (IDEaS) at Georgia Tech. Her research on randomized algorithms focuses on designing efficient sampling algorithms and her work has helped create an interdisciplinary field bridging computer science, discrete probability, and statistical physics. Randall is a Fellow of AMS and a national associate of The National Academies. Randall gave the AMS Arnold Ross Lecture to high school students and their teachers in 2009 and delivered AMS Invited Addresses at the Joint Mathematics Meetings in Baltimore in 2003 and 2018. She chaired the program committee for the Symposium on Discrete Mathematics in 2011 and was co-chair of the SIAM Conference on Discrete Mathematics in 2016. Previously, she was a DIMACS/IAS postdoctoral fellow and co-organizer of the DIMACS Special Focus on Discrete Random Systems.

Adversarial Risk Analysis (ARA) is a decision-theoretic alternative to game theory, applicable to corporate competition, auctions, and counterterrorism. In ARA, one builds a model for the strategic decision making of one's opponent(s), and then places subjective Bayesian distributions over unknown quantities. This structure enables the analyst to compartmentalize distinct kinds of uncertainty. Within this framework one can use standard Bayesian techniques to develop a probability distribution over the actions of the opponent. Given this distribution, the decision theorist chooses the action that maximizes expected utility.

Speaker Bio: David Banks is director of the Statistical and Applied Mathematical Sciences Institute (SAMSI) and Professor of the Practice in Statistical Science at Duke University. His research areas include models for dynamic networks, dynamic text networks, adversarial risk analysis (i.e., Bayesian behavioral game theory), human rights statistics, agent-based models, and forensics, as well as selected topics in high-dimensional data analysis. He is a Fellow of the Institute of Mathematical Statistics and of the American Statistical Association (ASA) and 2015 recipient of the ASA’s Founder’s Award. Banks is past-president of the Classification Society and has twice served on ASA’s Board of Directors. His book, Adversarial Risk Analysis, with David Ríos Insua and Jesus Rios, won the 2018 DeGroot Award. Banks was co-organizer of several DIMACS workshops, including the 2010 workshop on Adversarial Decision Making, and a leader in the DIMACS Special Focus on Computational and Mathematical Epidemiology.

Infectious salmon anemia (ISA) is a major threat to the viability of salmon populations, source of Omega-3 fatty acids, minerals and vitamins. ISA has caused high death rates in Atlantic salmon under marine farmed conditions. In this talk, we will introduce a discrete-time viral dispersal-linked farmed and wild salmon ISA disease model. Our ISA model includes salmon's pattern of cyclic dominance of subpopulations. We will compute the basic reproduction number, Râ‚€, and use it to explore the effects of viral dispersion on ISA invasion or eradication. In the absence of wild salmon, we will use simulations to show that higher proportion of viral dispersion from ISA infected farmed cages to the nearby body of water is capable of stabilizing cyclic enzootic ISA disease dynamics via period-doubling bifurcation reversal. In the event of an ISA outbreak in a farmed salmon population, we will illustrate that higher proportion of ISA viral dispersion from the infected farm could put the nearby wild salmon population at risk of the ISA disease.

Speaker Bio: Abdul-Aziz Yakubu is Professor of Mathematics at Howard University. His research interests lie in mathematical biology and dynamical systems and encompass theoretical investigations of nonlinear systems that arise in the diverse fields of ecology, population dynamics, epidemiology and demography. Yakubu has co-organized workshops held in association with the DIMACS Initiatives in Africa, was an organizer of the 2015 DIMACS Mathematics of Planet Earth workshop on Management of Natural Resources, and has been a leader in CCICADA, the homeland security center based at DIMACS. He led a team of student researchers visiting DIMACS for a summer, and he serves on CCICADA’s Executive Committee.

Too often, critical data are overlooked in the policy development process, and potentially compelling insights are missed that could drive new policies and programs. In addition, most public health analyses stop at suggesting solutions without providing an epidemiological or economic impact of their recommendations, and results and recommendations are not presented to decision-makers in an easily understandable format. This is especially true in low/middle-income countries (LMICs), and it is partially due to a shortage of skills to perform such analyses and present results. In 2015, the Bloomberg Philanthropies Data for Health Initiative started providing technical assistance in 20 LMICs to build their capacity to collect and analyze scientific data to guide and support health policy development and health program decision-making (https://www.bloomberg.org/program/public-health/data-health/). The Data to Policy (D2P) program comprises one component of this initiative, led by the U.S. Centers for Disease Control and Prevention and the New York-based nonprofit Vital Strategies. D2P trains Ministries of Health (MOHs) staff to use operations research and management science tools and thinking to analyze data to advance policy agendas and support specific policy priorities identified by governments through the development of data-driven policy briefs. Policy forums are held where briefs are presented to government officials and stakeholders so that discussions can start on addressing the public health issues. The final part of D2P is to make this training sustainable, where we teach MOHs staff to teach others, and thus build capacity. D2P has impacted twelve countries: Myanmar, Peru, Philippines, Solomon Island, Sri Lanka, Rwanda, Zambia, China, Tanzania, Bangladesh, Colombia, and Ghana. Of the 88 briefs generated by these countries, about 90% have been presented to decision makers, and 55% of them have had some influence in policy decisions. And four countries now have the capacity to teach D2P (Zambia, China, Colombia, and Myanmar).

Speaker Bio: Michael Washington is a Health Scientist at the Centers for Disease Control and Prevention, where he develops policy-based, quantitative models and analyses for use by Federal, State, and local public health authorities to advance the public health mission of the CDC. ​He is also building ministries of health capacities in many countries to create such models and do similar analyses.  Washington connected with DIMACS though the special focus on Mathematical and Computational Epidemiology and was heavily involved in DIMACS initiatives in Africa that brought U.S. and African researchers together to study a variety of issues related to epidemiology and health in Africa and around the world.

What are the contributions of genetics, environment, and stochasticity to cancer risk? We will present this in the context of a cancer predisposition syndrome where individuals born with a germline mutation in a tumor suppressor gene are at high risk for cancer including having multiple primary tumors in their lifetime. We will specifically address why related mutation carriers have different penetrance for cancer.

Speaker Bio: Chang Chan is Associate Professor of Medicine at Rutgers Robert Wood Johnson Medical School and co-leader of the Genome Instability and Cancer Genetics Program at Rutgers Cancer Institute of New Jersey, a DIMACS partner. Chan has developed and applied quantitative methods to the analysis of genomic, transcriptomic and epigenomic data. His work employs large datasets and models to bring insights into the biology of cancer. Chan has been a mentor in DIMACS REU Program.

The representation of women in computing is dismally low. In this talk, I will overview my experiences with programs focused on supporting participation of women in computing. In particular, I will describe a program developed while I was director of DIMACS, the Douglass-DIMACS-SAS Computer Science Living-Learning Community (LLC) for women at Rutgers, now in its fourth year. Each year, 20-35 first-year undergraduate women who intend to major in computer science (CS) apply and are selected to participate. LLC participants live in a common residence hall and are provided with an educational, mentoring, and community-building program that supports their progress as students and CS majors. Participants take a “house course,” Great Ideas and Insights in Computer Science, as a group, and also take a course on Knowledge and Power: Issues in Women’s Leadership. Program activities include study sessions and industry interactions, as well as opportunities to participate in K-12 outreach programs, hackathons, and computing research. To evaluate the program, participants and a similar comparison group are surveyed at the beginning and end of the academic year and a focus group is conducted with program participants. Program participants find the program valuable and would recommend it to others, but both program participants and the comparison group report some lack of confidence in their potential success as computer scientists. I will also briefly describe plans for a new program in Computer Science at Barnard College.

Speaker Bio: Rebecca Wright is the Druckenmiller Professor of Computer Science and the director of the Diana T. and P. Roy Vagelos Computational Science Center at Barnard College. Her research is primarily in the area of information security, including cryptography, privacy, foundations of computer security, and fault-tolerant distributed computing. Wright is a Distinguished Member of the ACM, a Fellow of the Institute of Electrical and Electronic Engineers, and the 2019 recipient of the SIGACT Distinguished Service Award for her “extraordinary service to the theoretical computer science community and beyond.” She is on the Board of Directors of the Computing Research Association's Committee on Widening Participation in Computing Research and led the development of programs at Rutgers designed to engage more women in computing. Prior to joining Barnard College, Wright served as Director of DIMACS, leading all of its various activities for over seven years.

Women remain highly underrepresented in the science, technology, engineering, and mathematics (STEM) workforce. Women hold 24 percent of STEM jobs in the United States, and only 2 percent of Latinas held science and engineering positions in 2015, according to the National Science Foundation. The dearth of women in STEM occupations is due to a complex interplay of several factors which include underrepresentation in STEM field majors, gender differences in employment expectations, and high rates of retention, to name a few. In this talk, I will describe some of the challenges that Latinas face when pursuing a STEM degree and employment. I will share my own experience in mathematics and biostatistics to draw parallel to the challenges that other Latinas face nationwide. I will describe how a career in STEM has shaped all aspects of my professional career through research, teaching, and service. I will present some preliminary findings from a recent study to illustrate an area of research involving health disparities.

Speaker Bio: Miriam Nuño is Associate Professor in Biostatistics and Surgery Residence at the University of California at Davis. Her research focuses on the application of statistical methods for analyses of electronic health records and claims data. It often emphasizes one of these three areas: comparative effectiveness research, causal inference in observational studies, and mathematical modeling for predicting patient outcomes. Nuño is the recipient of the faculty scholar award given by the Center for the Advancement of Multicultural Perspective in Science (CAMPOS). She serves in the Vice Chancellor’s Advisory Committee on Faculty Excellence in Diversity at UC Davis and is actively involved in efforts to increase the number of underrepresented minority students and faculty members in mathematics, biostatistics, and public health sciences. She supports the efforts to increase representation of women in STEM careers through the César Chávez Youth Leadership Conference and Celebration each year. Nuño first came to DIMACS as a graduate student to participate in a summer school tutorial on Dynamic Models of Epidemiological Problems in 2002 and later for a Working Group on Methodologies for Comparing Vaccination Strategies in 2004.

Deep learning has proved very successful at a host of AI/ML tasks, but theoretical understanding has lagged. The talk will survey ongoing efforts to understand the success of this method, including optimization aspects and a magical ability of vast nets to not overfit on tiny data sets.

Speaker Bio: Sanjeev Arora is Charles C. Fitzmorris Professor of Computer Science at Princeton University and Visiting Professor in Mathematics at the Institute for Advanced Study. He works on theoretical computer science and theoretical machine learning. He has received the Packard Fellowship (1997), Simons Investigator Award (2012), Gödel Prize (2001 and 2010), ACM Prize in Computing (2012), and the Fulkerson Prize in Discrete Math (2012). He is a fellow of the American Academy of Arts and Sciences and member of the National Academy of Science and was a plenary speaker at the International Congress of Mathematicians in 2018. He has participated in many DIMACS activities, including co-organizing the Special Focus on Hardness of Approximation.

Robotics is attracting significant interest given its potential impact and recent progress in mechanisms and sensors, as well as increased access to computational resources and data. At the core of the field, however, lie critical and long standing algorithmic challenges. This talk will highlight some of the problems in robotics of foundational importance as well as recent progress achieved. The challenges include how to compute high-quality paths for high-dimensional robotic systems as well as the combinatorial challenges that arise when multiple robots have to interact with multiple objects in an environment.

Speaker Bio: Kostas Bekris is an Associate Professor of Computer Science at Rutgers University. His research is in designing practically efficient algorithms with desirable formal guarantees in robotics, especially for motion planning and for coordination among multiple robots. He is recipient of an early CAREER Faculty Award by the National Aeronautics and Space Administration and currently leads an NSF-funded National Robotics Initiative project on learning in robotic manipulation with soft adaptive hands. Bekris joined Rutgers in 2012 and has been engaged with a variety of DIMACS and CCICADA projects, particularly those involving modeling and simulation of large numbers of physically-realistic agents in domains, such as transportation planning and security. He is also a leader of DATA-INSPIRE, a new transdisciplinary TRIPODS Institute based at DIMACS.

Twenty years ago, I was motivated by the Internet’s proliferation and Moore’s law to consider the architectural-challenges and application-opportunities of distributed sensor networks. The opportunity to bid for, build, and run the NSF Science and Technology Center for Embedded Networked Sensing (2000-2012) was personally the transformative experience of my professional life, and more importantly contributed to the shaping of many young researchers and research fields. CENS embraced societally important applications as the context for innovation, integrated theory and experimentation, and adapted to ride unanticipated technology trends -- the proliferation of mobile devices and machine learning. One of the key paradigms that emerged was Participatory Sensing, “data collection and interpretation in which individuals use their personal mobile devices and web services to systematically explore interesting aspects of their worlds ranging from health to culture,” which continues to pose technical and socio-technical challenges in the context of crowdsourcing, social media, IoT, mobile health, federated learning, edge computing, personal agents, personalization, and privacy.

Speaker Bio: Deborah Estrin is the Robert V. Tishman '37 Professor of Computer Science at Cornell Tech where she also serves as Associate Dean for Impact. Her research interests are at the intersection of user-centric data applications, personalization, and privacy. Her honors include the ACM Athena Lecture (2006), the Anita Borg Institute's Women of Vision Award for Innovation (2007), honorary doctorates from EPFL and Uppsala University, the IEEE Internet Award (2017), and MacArthur Fellow (2018). She is an elected member of the American Academy of Arts and Sciences (2005), the National Academy of Engineering (2007), and the National Academy of Medicine (2019). Before joining Cornell, Estrin was the Founding Director of the Center for Embedded Networked Sensing (CENS), an NSF Science & Technology Center based at UCLA.

This panel will engage experienced leaders of scientific centers in exploring the role of centers in the future. The panelists will address questions such as:

• How has the role of a center changed over the last 30 years and how do you see it changing in the future?
• How do centers add value to the scientific enterprise?
• Does the wide availability of virtual meeting capabilities change the role of today’s centers (or tomorrow’s)?
• How does the increasing need for multi-disciplinary teams to address complex problems affect the role of centers?
• How does the increasing emphasis on international collaborations suggest that centers should change?
• What makes a center sustainable?

Moderator: Rebecca Wright is emerita director of DIMACS and the Druckenmiller Professor of Computer Science and the Diana T. and P. Roy Vagelos Computational Science Center Director at Barnard College.

Panelists:

David Banks is director of the Statistical and Applied Mathematical Sciences Institute (SAMSI) and Professor of the Practice in Statistical Science at Duke University.

Deborah Estrin is the Robert V. Tishman '37

Three Decades of DIMACS is a unique opportunity to survey, in one lecture, subjects usually not treated together—Ramsey theory and structural graph theory, sparsity and complexity, and various algebraic and model theoretic connections. Just a few glimpses of rich and fast growing areas of contemporary mathematics and theoretical computer science.

Speaker Bio: Jarik Nesetril is Professor of Mathematics and Director of Institute of Theoretical Computer Science (ITI) at Charles University. He is also Director of the Center for Discrete Mathematics, Theoretical Computer Science, and its Applications (DIMATIA). His research interests include Ramsey theory, graph theory, homomorphisms, and constraint satisfaction. NešetÅ™il is a member of the Hungarian Academy of Sciences, the Academia Europaea, and a corresponding member of the German Academy of Sciences. In 2010, he received the Medal of Merit of the Czech Republic and was an invited speaker at the International Congress of Mathematicians in the sections on logic and combinatorics. One of the highlights of NešetÅ™il’s long interaction with DIMACS is the DIMACS/DIMATIA partnership in an international REU program that brings Czech students to DIMACS and sends U.S. students to DIMATIA each summer. 2019 marks the 20th anniversary of this pioneering partnership.

A graph is a split graph if its vertex set can be partitioned into a clique and an independent set. A split graph is unbalanced if there exist two such partitions that are distinct. We discuss a variety of results about balanced and unbalanced split graphs, including showing that these classes can be recognized by their degree sequences and that all threshold graphs are unbalanced split graphs. This is joint work with Karen Collins, Wesleyan University.

Speaker Bio: Ann Trenk is Professor of Mathematics at Wellesley College. Her research interests lie in graph theory and partially ordered sets, with a particular emphasis on structured families of graphs, including interval graphs, perfect graphs, and tolerance graphs. Her work in this area led to the book, Tolerance Graphs, which she coauthored with Martin Golumbic. Trenk has taught high school students both as a full-time teacher and in summer programs, organized math enrichment activities for elementary school children, and has led a variety of outreach activities for K-12 teachers. She spent a sabbatical year at DIMACS and has taught in a range of DIMACS summer programs, including the Young Scholars Program for high school students, Leadership Program for K–12 teachers, and DIMACS Connect Institute to engage high school teachers in research topics in discrete mathematics.

Much of the literature in fair machine learning takes the "statistical group fairness" approach: first fix a small collection of high-level groups defined by protected attributes (e.g., race or gender), and then ask for approximate parity of some statistic of the classifier (e.g., positive classification rate or false positive rate) across these groups. While this approach is amenable to tractable algorithmic formulations, it does not provide fairness protection on an individual level. In contrast, the "individual fairness" approach aims to address this limitation, but it relies on strong assumptions on the data (e.g., a known similarity measure across individuals). In this talk, I will cover some of my recent work that attempts to bridge the gap between the two fairness approaches and provides efficient algorithms that satisfy a range of fairness notions that interpolate between individual and group fairness.

Speaker Bio: Steven Wu is Assistant Professor of Computer Science and Engineering at the University of Minnesota. He is broadly interested in algorithms and machine learning, especially in the areas of privacy-preserving data analysis, fairness in machine learning, and algorithmic economics. He is recipient of the Facebook Mechanism Design for Social Good Research Award, as well as a Google Faculty Research Award and a J.P. Morgan Research Faculty Award. Wu was also a participant in the DIMACS Research Experiences for Undergraduates (REU) program in 2010.

The subject of mathematical ecology is one of the oldest and most exciting in mathematical biology, and has helped in the management of natural systems and infectious diseases. Though many problems remain in those areas, we face new challenges today in finding ways to cooperate in managing our Global Commons. From behavioral and evolutionary perspectives, our societies display conflict of purpose or fitness across levels, leading to game-theoretic problems in understanding how cooperation emerges in Nature, and how it might be realized in dealing with problems of the Global Commons. This lecture will attempt to weave these topics together, tracing the evolution from earlier work to challenges for the future.

Speaker Bio: Simon Levin is the James S. McDonnell Distinguished University Professor in Ecology and Evolutionary Biology at Princeton University. His principal interests lie in understanding how macroscopic patterns and processes are maintained at the level of ecosystems and the biosphere, in terms of mechanisms that operate primarily at the organism level, and in the interface between ecological and socio-economic systems. His work integrates empirical studies and mathematical modeling, with emphasis on how to extrapolate across scales of space, time, and organizational complexity. Levin is a member of the American Philosophical Society and the National Academy of Sciences, as well as Fellow of the American Academy of Arts and Sciences and the American Association for the Advancement of Science. His numerous honors include the National Medal of Science, Ramon Margalef Prize in Ecology and Environmental Sciences, Kyoto Prize, Tyler Prize, Heineken Prize and several honorary degrees. He has been organizer for many DIMACS activities, including the SF on Computational and Mathematical Epidemiology, Mathematics of Planet Earth, and DIMACS Initiatives on Epidemiological Modeling in Africa. He is a past member of the DIMACS Advisory Board.

Bio-Inspired design is an active area of engineering research, but most instances of successful application are case-by-case analogies constructed for isolated problems. In this talk, we will discuss how natural biological systems achieve robust and efficient solutions and the implications for helping inform how and when it may be most fruitful to construct these direct analogies to particular design problems. Critically, we will also discuss when instead we should probably forgo analogy and instead build tailored tools from scratch. We will consider how it might be possible to move beyond this case-by-case analogy paradigm to exploit the deeper design elements that make natural systems successful at converging on sufficient solutions and attempt to distill an abstract set of fundamental design tools that can be used in purposeful, original design.

Speaker Bio: Nina Fefferman is Professor of Mathematics and of Ecology and Evolutionary Biology at the University of Tennessee, where she also directs the Mathematical Modeling Consulting Center at the National Institute for Mathematical and Biological Synthesis (NIMBioS). Her research applies mathematical and computational models to biological systems, especially those created and governed by the voluntary collaboration of many independent individuals. Her work spans a number of different contexts, including epidemiology, evolutionary and behavioral ecology, and conservation biology. Fefferman has a long association with DIMACS, beginning as a visiting researcher in 2005 and continuing as a research faculty member through 2016 when she joined the University of Tennessee. She has organized many DIMACS events, led CCICADA projects, and been among our most popular REU mentors.

Education is a key facet of the NSF Science and Technology Center Program. STCs have developed novel educational programs for all educational levels and for students, teachers, faculty, administrators, and professionals wanting additional education. This panel will highlight the impacts of and issues in developing and implementing pioneering education programs. DIMACS educational programs such as REU, programs developing modules for high school classrooms, and Reconnect workshops for college faculty have been effective at integrating research and education and in engaging diverse participants but need rethinking as the fundamental tools and goals of education change. Reconnect workshops expose faculty teaching undergraduates to current research topics relevant to the classroom.  DIMACS’s long-running REU program immerses undergraduates in a multidisciplinary environment that exposes them to a broad range of computational and mathematical topics applied in contexts that range from bioinformatics to big data. Notable features include an international component coordinated with Charles University in Prague, which established the first NSF-supported international REU program. The REU partnership between DIMACS and Charles University has now endured for twenty years. As we look to address global challenges, we need to think about new ways to better equip students to collaborate across boundaries—both between disciplines and between nations.

For more than a decade, DIMACS has led efforts to develop novel materials that bring mathematical and computational methods and reasoning into high school classrooms. These materials take the form of short “modules” that can be flexibly adapted for use in a variety of courses at a variety of grade levels. The interdisciplinary nature of the modules offers students a contemporary view of science as a multidisciplinary enterprise. Modules in computational thinking are now forming the basis for a new DIMACS project developing and testing an online course for teacher professional development in computational thinking. What would be required to develop modules on socially responsible algorithms? On intelligent machines and their interaction with humans? On new domains for cybersecurity? On other topics yet to be determined?

This panel will discuss these and other topics, including the influence DIMACS has had on their lives and careers, on curricula in K-16, and on professional development. They will also provide examples of where DIMACS can undertake new novel educational programs.

Moderator: Margaret Cozzens is DIMACS Associate Director for Education. She served as Division Director of Elementary, Secondary, and Informal Education at the National Science Foundation for seven years (1991–1998) and has developed and organized many of the most unique and pioneering DIMACS programs.

Panelists:

Kristina Adams is Assistant Professor and Coordinator of Secondary Education at Cottey College. Before joining Cottey, Adams taught science in public schools for 18 years, provided professional development to teachers across the state of Oklahoma, and was the secondary education coordinator as well as a math and science professor at St. Gregory’s University.  Adams came to DIMACS in her role as a science teacher, testing materials integrating biology and mathematics developed in DIMACS-led projects with her students. She is now one of the site leaders for a DIMACS project providing on-line professional development to enable teachers to bring computational thinking into their classrooms.

Sol Garfunkel is Director of COMAP, the Consortium for Mathematics and Its Applications, an award-winning non-profit organization whose mission is to improve mathematics education for students of all ages. Among its many activities, COMAP holds the annual Mathematical Contest in Modeling (MCM), an international contest for high school students and college undergraduates. Garfunkel has worked with DIMACS on numerous projects developing materials in mathematical biology, computational thinking, and sustainability for high school classrooms.

James Kupetz is a science and mathematics teacher at Pittston High School in Pittston, PA. In his long association with DIMACS, Kupetz has participated in virtually every major DIMACS K-12 education program. He was a participant in early DIMACS programs such as the DIMACS Connect Institute and is an author of DIMACS modules for high school classrooms in bio-math, computational thinking, and sustainability. He is a frequent Reconnect participant and is currently one of the site leaders for the DIMACS project providing on-line professional development for teachers in computational thinking.

Martin Loebl is Professor of Applied Mathematics at Charles University in Prague. Loebl’s association with DIMACS began when he was a postdoc and has continued through the long-standing partnership between DIMACS and the DIMATIA Center at Charles University. The two centers have collaborated to run an international REU program for two decades. At present, Czech participation is part of the Combinatorial Structures and Processes (CoSP) project coordinated by Loebl and funded through the European Union’s Horizon 2020 research and innovation programme.

Lara Pudwell is Associate Professor of Mathematics and Statistics at Valparaiso University. As a graduate student at Rutgers, Pudwell was graduate student coordinator for the DIMACS REU program for three years running. As a faculty member, she remains committed to undergraduate research, having directed her own REU program and coauthored the recent book, A Mathematician’s Practical Guide to Mentoring Undergraduate Research, to help others do the same.

Martene Stanberry is Associate Professor of Mathematical Sciences at Tennessee State University. As an applied mathematician, she is interested in developing new courses and undergraduate curricula related to applied mathematics and connecting mathematics with other areas of STEM. She received a DHS Scientific Leadership Award for Minority Serving Institutions to collaborate with CCICADA and later participated in several Reconnect workshops and DIMACS activities related to the Mathematics of Planet Earth.

This panel explores the potential for mathematical and computational research to enhance safety and security.

DIMACS has been home to two centers for research related to homeland security—first DyDAn (established in 2006) and now CCICADA (established in 2009). These centers fueled research on homeland security topics and were engines for technology transfer and new education programs. Security-related projects through these centers have dealt with the role of technology in cyber forensics, screening at airports and sports stadiums, maritime cyber security, allocation of resources to locations or missions, customs fraud and smuggling, analytics in law enforcement, privacy and civil liberties, spread of infectious disease, voice forensics and hoax calls, and human trafficking, among others. Many of these projects led to transfer of technology and are revealing new mathematical and computational challenges for the future. The formation of DyDAn and CCICADA broadened both the research community and the funding sources sustaining DIMACS—two factors that are critical to keeping a center vibrant and sustainable. This panel explores the value of such models for DIMACS and other centers in the future.

Moderator: Dennis Egan is Assistant Director for CCICADA, the homeland security center based at DIMACS. He has a long history with DIMACS, stemming from the days when he was the representative of partner Telcordia Technologies to the DIMACS Executive Committee.

Panelists:

Daniel DeLorenzi is Vice President for Security and Safety Services at MetLife Stadium. He has been heavily involved in collaborating with CCICADA in programs to enhance security at sports stadiums, has provided a laboratory for CCICADA faculty and students to explore new ideas, and participated in numerous CCICADA events and research projects.  

Bradford Greening is a health scientist and mathematical modeler at the Centers for Disease Control and Prevention where he works to assess, develop, and improve emergency preparedness and response efforts in a wide range of public health applications including influenza, Ebola, Zika, and nuclear/radiation. Greening first came to DIMACS as an undergraduate participant in the 2010 REU program where he was mentored by Nina Fefferman, who later became his Ph.D. dissertation advisor. As a graduate student, Greening received a three-year fellowship through CCICADA and participated in a wide range of DIMACS and CCICADA activities. 

Paul Kantor is Emeritus Professor of Library and Information Science at Rutgers University. His long association with DIMACS began in earnest after the 9/11 attacks, resulting in a series of grants from the intelligence community’s KDD program, as well as projects on inspection at ports of entry and detection of nuclear materials. Kantor has been heavily involved with CCICADA and its predecessor, DyDAn, and served as research director for CCICADA.

Asamoah Nkwanta is Professor of Mathematics at Morgan State University. His association with DIMACS began through the Reconnect program which led to long-term engagement with DyDAn and CCICADA. In association with DyDAn, Nkwanta and two students spent a summer at DIMACS working on problems in epidemiology and traveled to participate in related DIMACS events in Africa.

Warren Powell is Professor of Operations Research and Financial Engineering at Princeton University. Powell and his students became involved with DIMACS through applications in nuclear detection and other areas of homeland security that involved learning under uncertainty. He participated in both CCICADA and its predecessor, DyDAn, and he led efforts in optimal learning in the context of nuclear detection.

William (Al) Wallace is Professor of Industrial and Systems Engineering at Rensselaer Polytechnic Institute. A long-time member of DIMACS, Wallace was the lead for RPI’s partnership in CCICADA and DyDAn, bringing expertise in emergency response and management and resilience planning.

Given a collection of points, the TSP asks for the shortest route to visit them all. Simple enough. But even a whisper of the problem strikes fear in the heart of the computing world. Last year, a Washington Post article reported it would take "1,000 years to compute the most efficient route between 22 cities." This claim, however, ignores 70 years of intense study. A 22-city TSP can be handled in a snap with modern algorithms, even on an iPhone. Going larger, we describe techniques that have been used to solve to precise optimality examples with nearly 50,000 points and Google Map walking distances. And if you need to visit the nearest 2,079,471 stars, there is a route, ready to go, that is guaranteed to be no more than 1.000009 times longer than a shortest possible tour.

This work follows a long line of computational research going back to Julia Robinson in 1949 and Dantzig, Fulkerson, and Johnson in 1954. We will discuss the history of the TSP and its applications, together with recent computational efforts towards exact and approximate solution methods. The talk is based on joint work the David Applegate, Daniel Espinoza, Marcos Goycoolea, and Keld Helsgaun.

Speaker Bio: Bill Cook is Professor of Applied Mathematics and Statistics at Johns Hopkins University. His research interests lie in integer programming and combinatorial optimization, where he is especially known for his work on the Traveling Salesman Problem (TSP). Cook is a member of the National Academy of Engineering and a Fellow of the AMS, INFORMS, and SIAM. He is a recipient of the Beale-Orchard-Hays Prize of Mathematical Programming Society and the Lanchester Prize for his book (with Applegate, Bixby, and Chvatál) The Traveling Salesman Problem: A Computational Study. In 1998, he was an invited speaker at the International Congress of Mathematicians. Cook was an organizer of the DIMACS Special Year on Combinatorial Optimization and has taken part in DIMACS Implementation Challenges, both as an organizer (of the 2nd Challenge on NP Hard Problems: Maximum Clique, Graph Coloring, and Satisfiability) and a competitor (in the 8th Challenge on the Traveling Salesman Problem).

Some years ago, I was an REU student at DIMACS, where I met Dieter van Melkebeek and Lance Fortnow. At the time, they had recently proved some time-space tradeoff lower bounds for satisfiability (building on recent work of Lipton and Viglas). I will discuss this topic, some of my own later contributions to it, and its role in inspiring the circuit lower bounds program that currently drives my group's work.

Speaker Bio: Ryan Williams is Associate Professor of Electrical Engineering and Computer Science at MIT. His research is in algorithm