|Abstracts - Plenary Speakers|
We analyze an interactive model of credit ratings where external shocks, initially affecting only a small number of firms, spread by a contagious chain reaction to the entire economy. Counterparty relationships along with discrete adjustments of credit ratings generate a transition mechanism that allows the financial distress of one firm to spill over to its business partners. Such a contagious infectious of financial distress constitutes a source of intrinsic risk for large portfolios of credit sensitive securities that cannot be ``diversified away.'' We provide a complete characterization of the fluctuations of credit ratings in large economies when adjustments follow a threshold rule. We also analyze the effects of downgrading cascades on aggregate losses of credit portfolios. We show that the loss distribution has a power-law tail if the interaction between different companies is strong enough.
This paper reviews general theories for introducing liquidity into asset pricing and trading models.
Network models may be applied to many complex systems, e.g. the Internet, the World Wide Web, inter-bank lending/borrowing networks, etc. Cascade dynamics can occur when the (binary) state of a node is affected by the states of its neighbours in the network, for example when the default of a bank causes some of its creditors to default in turn. Such models have been used to aid understanding of the spread of cultural fads and the diffusion of innovations (Watts 2002), and can be generalized to include a variety of other cascading dynamics on networks (Gleeson 2008). I will review the techniques used to study these dynamics and examine their applicability for the modelling of contagion and systemic risk within banking networks (Nier et al. 2007, Gai and Kapadia 2007).
Network Security & Cryptography
In this presentation we review the model for related-key attacks and discuss on its significance. Then, we review recent results to identify which standards are insecure.
Better Architectures and New Security Applications for Coarse Network Monitoring (Mike Reiter University of North Carolina at Chapel Hill)
The smart phone revolution has fundamentally changed the way that people interact with information. However, this transition has not been marked by vigilant analysis of security and privacy risks. A consequence of this failure is that our online lives are arguably more at risk than any other time in history. This talk explores ongoing research, key challenges, and cautionary tales in cell-phone security. Focusing primarily on the Android operating system, we describe the limitations of existing security frameworks, and posit alternate designs tailored to address the often complex and sometimes conflicting requirements of the diverse stakeholders, i.e., users, application developers, handset manufacturers, and cellular carriers. Three policy frameworks--Kirin, Saint, and Porscha are described and experiences supporting practical applications and environments provided. The talk concludes with a brief discussion of the technical and economic challenges inherent in contemporary application markets.
As part of a push to improve network security, an increasing proportion of Internet traffic is being protected by end-to-end encryption. Though helpful for preventing eavesdropping and other attacks, encryption presents a challenge for network monitoring. In particular, attackers are using encryption to hide their activities from network intrusion detection systems.
It turns out, however, that encryption does not hide all features of communication; in particular, side information, such as packet headers and timings or connection statistics are available to observers. This information can weaken the privacy protections provided by encryption, and at the same time provide network monitors with insight into potentially malicious activity. I will present a short survey of research results on both sides of this problem and discuss some of our recent work on detecting coordinated attacks, such as peer-to-peer botnets, using encrypted traffic analysis.
Current large-scale information systems involve rich patterns of social interaction, and understanding the network structure of these interactions is crucial both for the design of the systems and for our understanding of human social dynamics at global scales. Social networks in many on-line settings encode a mixture of positive (friendly) and negative (antagonistic) relationships, but the bulk of research on these networks to date has focused almost exclusively on the positive interpretations of the links. We discuss how the interplay between positive and negative relationships affects the overall functioning of on-line social networks, and we connect our analysis to theories of signed networks from social psychology. We find that the signs of links in the underlying social networks can be predicted with high accuracy, using models that generalize across a diverse range of applications. In addition to providing a perspective for reasoning about the underlying networks and applications, this analysis provides one of the first large-scale evaluations of these theories using on-line datasets. We find that classical theories capture certain of the underlying effects, but that they are also at odds with some of the fundamental phenomena we observe--particularly related to the evolving, directed nature of the networks.
Public Health Issues in the Adult Film Industry: Report on an HIV Outbreak and the Current Status of STDs among Performers, 2004-2008 (Peter R. Kerndt, MD, MPH.Los Angeles County Department of Public Health)
Background: Production of adult films is a legal, multi-billion dollar industry in California. It is estimated that 200 adult film production companies in Los Angeles County employ 6,000 workers, of whom 1,200 are performers engaging in direct occupationally related sexual contact. In April 2004, the Los Angeles County Department Public Health (LAC DPH) received reports of three incidents of workplace-related HIV transmission in this industry, involving one male index case and three female performers.
Methods: LAC DPH and Cal/OSHA initiated investigations regarding these HIV transmission events. LAC DPH also sought technical assistance from NIOSH and the CDC to develop workplace health and safety recommendations for the adult film industry and established ongoing surveillance for STDs among performers.
Results: The index case had been regularly tested for HIV by PCR and had tested negative in February and March 2004, but subsequently tested HIV-positive in April. Between his two negative tests, the index case performed in film productions in Brazil, engaging in unprotected vaginal and anal intercourse followed by a short flu-like illness. Upon returning to California, the index case performed in adult film productions with 13 female partners, with whom he engaged in unprotected vaginal, and anal intercourse. Three of these female performers subsequently tested HIV-positive (23% attack rate); all had tested HIV-negative by PCR within the preceding 30 days. Nearly all other performers with whom these three women had sexual contact 30 days prior to their diagnoses were subsequently tested; none were HIV-positive. DNA sequencing has shown that the p17 region of gag and the gp41 region of env in blood specimens from the index case and two of the HIV-positive female performers were 100% similar, supporting the conclusion that the index case transmitted HIV to both women through occupational exposure. Surveillance in the period 2004-2008, identified over 3,200 chlamydia and gonorrhea cases among performers; ~75% were in female performers and ~25% were reinfected within one year.
Conclusions: Current workplace practices in the adult film industry, including lack of condom use, create substantial risk for transmission and acquisition of HIV and other STDs. Reliance on testing alone for prevention is insufficient to prevent workplace HIV/STD transmission. Transmission to others not employed in this industry is not fully understood. Mapping of sexual networks is needed to target public health interventions.
Presentation Slides [PDF, 1.2MB]
A major paradigm shift has happened in North American society these days:
Graphs resulting from human behavior (the web graph, social networks, etc.) have hitherto been viewed as a monolithic class with similar characteristics. Our focus here is on the compressibility of such graphs. It has been empirically shown that Web graphs can be compressed down to three bits of storage per edge. In light of this, we address two basic questions: are social networks as compressible as Web graphs and are there tractable and realistic models that yield highly compressible graphs?
Presentations slides [PDF, 800kB]
Online social networks have revolutionized the way we interact and share information over the Internet. Popular social networking applications include YouTube, Flickr, MySpace, Facebook, and Twitter, which support millions of active users. While being enormously popular, these applications only scratch the surface of what is possible to do, and there are tremendous opportunities in developing new, more advanced online social networking applications. Creating such applications poses new and fascinating research problems. One of major research challenges in this domain is to develop a formal understanding of online social networks both in terms of how online social networks are formed, and how they can be used to efficiently share and distribute information. In the talk, we will discuss research aiming at creating a mathematical foundation of online social networks that provides a formal understanding and framework for the design and analysis of algorithms for online social networking applications. The first part of the talk will present a broader research agenda for online social network. The second part will focus on recent theoretical results on search algorithms in online social networks. An interesting aspect of the results that we obtain is that they provide insight into why real-life social networks are so efficient.
Scale-free networks are often used to model HIV transmission networks such as social networks of injecting drug users and sexual networks. In both types of networks, person-to-person influences play an important role in propagating risk behaviour. HIV transmission in the network is then strongly influenced by this risk behaviour. We use a stochastic model on scale-free networks to describe both propagation of risk behaviour and HIV transmission.
The model is analysed through both mean-field methods and simulation studies. The two most important strategies for controlling the HIV epidemic are harm reduction programmes, which seek to reduce risk behaviour, and treatment as prevention programmes with highly active antiretroviral therapy, which reduce the infectiousness of people who are HIV positive. The network model is used to evaluate the potential effectiveness of both types of strategies for controlling the HIV epidemic.
Presentation Slides [PDF, 1.3MB]