Abstracts - Short Courses |

## Financial Networks and Risk Assessment
Topics and questions being addressed include: - What is systemic risk?
- What is the basic economic picture of the financial system as a random graph?
- What does the theory of ``small world'' random graphs hint about the nature of the financial system?
- What conclusions have resulted from various economic studies of the financial systems of various countries?
- What important effects did pre-crisis models miss that we can now learn from?
- What can mathematical finance say about systemic risk?
- Liquidity risk: how can it be better accounted for?
For more information, click here.
Topics to be addressed include: - Systemic risk
- Network models of banking systems
- Default contagion and systemic risk in counterparty networks
- Monitoring systemic risk
- Credit default swaps and systemic risk
For more information, click here.
## Network Security & Cryptography
Learn and become familiar with a combine hardware and software technology with which you can virtually intercept and decode any wireless signal. In this tutorial, you’ll learn the hardware platforms for Software Defined Radio (SDR). The difference with software-controlled radio will be clarified. The software architecture of a SDR will be explained. SDR-based signal interception cases will be reviewed. The tutorial puts an emphasis on the basic concepts of digital signal processing required for an understanding of SDR, with several signal-decoding examples. This is a beginner’s tutorial with active participation. There will be a lot of hands-on exercises done by the participants to put in practice the presented concepts. The Octave or Matlab software is used as a teaching aid. Each participant must come with a laptop with the Octave software installed (most of the systems are supported; free download from: http://www.gnu.org/software/octave/) or the Matlab software installed.
Ubiquitous computing systems employ small, inexpensive, interlinked devices distributed at all scales and deployed for a variety of applications, extending the traditional, already decentralized, network paradigm to an entirely pervasive setting. The security of such systems can only be as good as their weakest link. In this course we focus on RFID applications. Starting with the EPCGen2 standard we shall consider applications ranging from single tag interrogations to group interrogations. Security will be discussed in a practical framework that supports modularity and composability (maintained by refreshment).
Elliptic curve cryptography has seen wide spread adoption in systems requiring a high-level of security in constrained environments. A variety of ECC techniques have been engineered and deployed to meet these demands. A general primer on elliptic curve cryptography will be provided followed by a walkthrough of some unique applications, the evolution of techniques to meet these demands and an examination of the security properties of the deployed techniques.
In this lecture we will review the state of the art for symmetric cryptographic algorithms. We will focus on the challenges that we face to improve the tradeoffs between long term security, low footprint (and cost), and high performance. As examples, we will use the hash function crisis (the 2004-2005 attacks on MD4, MD5, and SHA-1), the progress in block ciphers and the status of high speed stream ciphers and MAC algorithms. The focus will not be on the design internals of cryptographic algorithms, but on how to select the most suitable algorithm for particular constraints (e.g., high performance versus lightweight) and on secure implementations.
Is it possible to carry out a secret key operation in software (e.g. to do an AES decrypt or RSA sign operation) without exposing the secret key to an adversary who has full visibility and control over the executing code? This is central question that motivates research in White-box Cryptography, and also motivates many vendors to consider Cloakware technology. ## Social Networks
As social creatures, our online lives just like our offline lives are intertwined with others within a wide variety of social networks. Each reply to an email, link to a blog, or a comment on an online video explicitly or implicitly connects one online participant to another and contributes to the formation of various social networks. Once discovered, these social networks can provide us with an effective mechanism for identifying and studying collaborative processes within online communities. In addition to being useful for researchers, information about online social networks can also be useful to people from a wide variety of other areas. For example, web developers can use information about online social networks to improve recommendation systems by analyzing the preferences of other users with similar interests. And companies can use online social networks to fine tune their recruitment efforts, organize more effective virtual marking campaigns or build brand loyalty using customer networks. For more information, click here. Presentation Slides [PDF, 1.9MB]
Complex networks arise in many diverse contexts, ranging from web pages and their links, protein-protein interaction networks, and social networks. The modelling and mining of these large-scale, self-organizing systems is a broad effort spanning many disciplines. A number of common properties have been observed in complex networks, such as power law degree distributions and the small world property. New stochastic graph models simulate these properties, while expanding our theoretical understanding of random graph models. Models for complex networks also give insight into their underlying generative properties. With the current popularity of on-line social networks (or OSNs) such as Facebook, LinkedIn, and Twitter, there is an increasing interest in their measurement and modelling. In addition to other complex networks properties, OSNs exhibit shrinking distances over time, increasing average degree, and bad spectral expansion. Unlike other complex networks such as the web graph, there are relatively few models for OSNs. In on-line social networks, models may help detect and classify communities, and better clarify how news and gossip is spread in social networks. We will give an overview of complex networks in the first half of the short course, including their properties and models. In the final half, we will consider models of OSNs. We will discuss new geometric models that suggest a reverse engineering approach: given only the graph structure, use the model to help uncover the hidden reality of the network. Presentation Slides [PDF, 7.8MB]
One application is that, if one has an organization such as a government office, how should one structure it so that, if workers randomly succumb to the flu, the office will, with the greatest probability, still be able to function? REFERENCES - Heski Bar-Isaac and Mariagiovanna Baccara, "How to Organize Crime," Review of Economic Studies 75 (2008), 1039–1067.
- Victor Campos, Vasek Chvatal, Luc Devroye, and Perouz Taslakian, "Transversals in Trees," unpublished.
- Walter Enders and Xuejuan Su, "Rational Terrorists and Optimal Network Structure," Journal of Conflict Resolution 51 (2007), 33-57.
- Jonathan David Farley, "Breaking Al Qaeda Cells: A Mathematical Analysis of Counterterrorism Operations (A Guide for Risk Assessment and Decision Making)," Studies in Conflict and Terrorism 26 (2003), 399-411.
- Jonathan David Farley, Toward a Mathematical Theory of Counterterrorism: Building the Perfect Terrorist Cell (U.S. Army War College, Carlisle Barracks, Pennsylvania, 2007).
- G. Gunther and B. L. Hartnell, "On Minimizing the Effects of Betrayals in a Resistance Movement," Proceedings of the Eighth Manitoba Conference on Numerical Mathematics and Computing (1979), pp. 285-306.
- R. Lindelauf, P. Borm, and H. Hamers, "The Influence of Secrecy on the Communication Structure of Covert Networks," Social Networks 31 (2009), 126-137.
- Nasrullah Memon, Jonathan David Farley, David L. Hicks, and Torben Rosenørn, Mathematical Methods in Counterterrorism (Springer Verlag, Vienna, 2009).
In many cases the only practical way to obtain a large enough sample from the population is to follow links from sample individuals to add more individuals to the sample. For example, in studies of the risk behaviors in people at risk for HIV/AIDS, the population is hidden and not listed on any frame so that standard sampling designs can not be applied. Instead, researchers follow social referrals from individuals in the sample to find more members of the hidden population. Similarly, in studies of the World Wide Web, links or connections from sites in the sample are followed to add more sites to the sample. Network methods also turn out to be useful for spatial sampling in environmental and ecological sciences where the populations tend to be highly clustered or rare. A variety of link-tracing sampling designs will be described, together with design-based and Bayes methods for estimating population characteristics based on such samples. Computational methods and available software will also be described in this course. |