ICCS 2011 kicked off yesterday and will continue until friday this week.  High-profile speakers that I’m pumped to see include Gene Stanley (statistical physics, econophysics), David Gondek (IBM Watson), Alex Pentland (human dynamics, reality mining), Geoffrey West (cities and scaling), Stephen Wolfram (Mathematica, Alpha), and, the one I’m most excited about, Nassim Taleb.  Here is the list of poster presenters and speakers.

I will be presenting a talk today and a poster tomorrow, both on research I have been doing at NECSI.  Here are the abstracts of the talk and poster, respectively.

Quantifying Multi-Scale Structure and Capabilities in Complex Systems

A central challenge in building and managing human organizations is identifying the proper structure to adopt. This problem affects all tasks ranging from those of small teams to global organizations. Today, costly trial and error learning dominates reorganization efforts. Here we characterize the coordination structure of an organization in order to match the structure to a similar characterization of the task to be performed. This characterization involves both the scale of tasks (the number of individuals or other organizational components needed to perform a task) and the variety of tasks that must be performed.

We quantify the scale and variety dependent capability using the complexity profile, and generalize Ashby’s Law of Requisite Variety to describe the matching of system organization to task. A successful organization optimizes tradeoffs between variety and scale to match the complexity profile of its tasks.

Here we develop a quantitative method and computational tools to calculate the multi-scale structure of the system and a set of tasks. We demonstrate through exam- ples that organizing a system effectively for a task requires matching both the scale and complexity of the organization to the scale and complexity of the task. These results provide a framework to calculate the organizational structure for successfully performing specific tasks.

Applications of our approach are relevant to a wide range of systems. Challenges in the economic, financial, development, education, and military sectors highlight that many of the most difficult problems stem from inability to meet multi-scale demands. Characterizing the structure of successful solutions given a task can provide direct insight into how to address these problems.

Characterizing Complex Terrains: Mathematical Foundations and Applications

We generalize fractal and affine analysis to describe the multi-scale structure of systems that are not self-similar. The approach is based upon the complexity profile and, unlike multifractal analysis, directly maps the scale dependence of key properties of the structure. This technique is of both fundamental and practical importance in a wide range of real world contexts. We apply this method to construct a natural characterization of scale and complexity of a given geophysical terrain. We develop an approximation that is useful for high resolution elevation data, and calculate the complexity profiles for Afghanistan and Iraq. The analysis provides direct insight into the human social dynamics, including conflict, that occur in these countries.