SASO2010

SASO 2010

Fourth IEEE International Conference on Self-Adaptive and Self-Organizing Systems

Budapest, Hungary, September 27-October 1, 2010

Contact

E-mail address of contact person as image

Search

Loading

Sponsors

Invited Talks

Cooperation, Norms, and Conflict in Peer-to-Peer and Social Systems

September 29th, Wednesday

Dirk Helbing

Invited speaker: Dirk Helbing (ETH Zurich, Zurich, Switzerland)

Abstract: Peer-to-peer systems are distributed systems, the functioning and performance of which implictly requires coordination and cooperation between peers. The problem has a lot in common with the creation of public goods in social and economic systems. Therefore, we will study, what are mechanisms, which support spontaneous cooperation and the development of commonly shared behaviors ("norms"). In this connection, it is important to study the role of social mechanisms such as repeated interactions, group selection, community formation, costly punishment and group pressure, and how they allow to transform social dilemmas into interactive situations that promote collaboration and performance. It is also discussed, under what conditions social cooperation or system performance can suddenly break down. Moreover, when people stop agreeing on common norms, this is likely to produce conflict. These issues will be systematically studied within the framework of evolutionary game theoretical models. It is surprising that many aspects of complex collaboration systems, such as the outbreak and breakdown of cooperation, the formation of norms or communities, and the occurence of conflicts can be described by a unified theoretical framework.

Short bio: Since June 1st, 2007, Dirk Helbing (born on January 19, 1965) is Professor of Sociology, in particular of Modeling and Simulation at ETH Zurich. Before, he worked as Managing Director of the Institute for Transport & Economics at Dresden University of Technology, where he was appointed full professor in 2000. Having studied Physics and Mathematics in Göttingen, his master thesis dealt with the nonlinear modelling and multi-agent simulation of observed self-organization phenomena in pedestrian crowds.
Two years later, he finished his Ph.D. at Stuttgart University on modelling social interaction processes by means of game-theoretical approaches, stochastic methods and complex systems theory, which was awarded two research prizes. After having completed his habilitation on traffic dynamics and optimization in 1996, he received a Heisenberg scholarship. Both theses were printed by international publishers. Apart from this, Helbing has (co-)organized several international conferences and (co-)edited proceedings or special issues on material flows in networks and cooperative dynamics in socio-economic and traffic systems. He has given 250 talks and published more than 200 papers, including several contributions to journals like Nature, Science, or PNAS, which were discussed by the public media (newspapers, radio, and TV) more than 200 times. He collaborates closely with international scientists. For example, he worked at the Weizmann Institute in Israel, at Xerox PARC in Silicon Valley, at INRETS in Paris and the Collegium Budapest - Institute for Advanced Study in Hungary, where he is now a member of the external faculty.

Presentation: (pdf) 13M

Computing in a Dangerous World

October 1st, Friday

Howard Shrobe

Invited speaker: Howard Shrobe (MIT Computer Science and Artificial Intelligence Laboratory, Cambridge, USA; DARPA TCTO, Arlington, USA)

Abstract: We all know that cyber-security has become a major concern across the world. The number of attacks, the sophistication of the attackers and the monetary damage have all been increasing at exponential rates for several years. Our current approach can best be characterized as P4: Perimeter Protection, Patch and Pray. But this is clearly not working, at a recent "hackers vs software" contest all of the most popular software systems were penetrated within the first day. This isn't surprising, we see the same results at all such contests. But it is shocking that all of the major vendors have invested significantly in defenses against just such attacks, apparently to little effect. What is truly frightening is that this very software, or very similar software, that runs the critical infrastructures of all major countries. Any of those infrastructures could be compromised at any time by attackers with appropriate motivation and skill.
We need ways to make our computer systems resilient in the face of such attacks and this involves, in my view, completely new ways of designing computer systems (both hardware and software) that lead to systems that are in crucial ways adopt the strategies of organisms and societies. The immune systems of higher organisms for example include "innate" elements that are fast and deadly, but deal with a fixed set of pathogens that are always in the environment. But they also include a second "adaptive" system that responds more slowly but can mount adaptive responses to novel pathogens. Moreover, these systems interact. Such biological systems invest enormous resources into self-defense at a level that any self-respecting computer designer would regard as untenable. But I will argue that we have reached a point where the trade-offs for computers are more like those for biological systems, resources are abundant and the lack of adaptive self-defense can be fatal.
Finally, I will describe the next crisis that is coming. As aggressive scaling takes micro-electronics to ever finer geometry, we are rapidly approaching a time when devices will not work like "ideal" and "perfect" switches but will rather begin to be unreliable in ways that we have not had to cope with since computers were built from vacuum tubes. These "attacks from nature", transient errors, high device failure rates, are different from cyber-attacks; cyber attacks are conscious attacks on specific targets that have have value to the attacker. Nature is just "out to get everything". Thus if we are to continue to reap the benefits of Moore's law, we will have to begin to design processors to be resilient to device failure. In both cases, the standard tools of self-adaptive computation will prove to be key: self-monitoring, diagnosis and repair.

Short bio: Howard Shrobe received his BS from Yale College in 1968 in Mathematics and his MS and PhD in Computer Science from MIT in 1975 and 1978 respectively. While a student, he started the Programmer's Apprentice project with Charles Rich and Richard Waters. He was also involved in the design of the Scheme '81 microprocessor at MIT. He joined Symbolics Inc in 1982 where he help to design the Ivory microprocessor. In 1994, Dr. Shrobe joined the Software and Intelligent Systems Technology Office at DARPA and later become Chief Scientist of the Information Technology Office. In that role, he launched DARPA's Information Survivability program. Upon returning to MIT in 1997, Dr. Shrobe became Associate Director of the Artificial Intelligence Laboratory and leader of the Intelligent Room research group. In February of 2010, Dr. Shrobe joined the Transformational Convergence Technology Office (TCTO) of DARPA where he is focusing on starting new research effort on resilient systems for cyber security. The first of these CRASH (Clean-slate design of Resilient Adaptive Secure Hosts) had just begun.

Presentation: (pdf) 3M