From Santa Fe Institute Events Wiki

Revision as of 15:07, 21 April 2015 by Keith.umd (talk | contribs) (Providing profile information)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

What are your main interests?

Keith's main interests fall under the theme of "information propagation", especially competing ideas and opinions. More specifically, his interests relate to determining how opinions compete within a social network, as well as why a small set of ideas may eventually dominate (e.g., the relative success of certain political parties).

What sort of expertise can you bring to the group?

Keith is knowledgeable in statistical mechanics, including scaling laws and other critical phenomena. Critical phenomena is commonly seen in complex systems not often associated with physics, such as the nervous system, or opinion dynamics. Therefore, it is important to learn about critical phenoma in the hopes of better understanding the behavior of these systems.

For example, recent literature has suggested that the dynamics of the brain falls into the Critical Branching Process Universality Class, plausibly explaining the wide variance of neural cascade sizes and times before a cascade ends. Of course, in this example, the question then goes one layer deeper: why does our brain seem to self-organize into this universality class? This is a difficult question with so far few satisfactory answers.

Another example, more along the lines of Keith's current work, involves opinion dynamics. Some of his current work is on modeling competing opinions, where it is observed that model parameters of opinion dynamics tend to be close to critical values in order to agree with empirical data. In this particular regime, one sees critical slowing down and scaling laws that appear, which are apparent in empirical data as well.

What do you hope to get out of the CSSS?

Keith hopes to learn more about seemingly disparate fields with the goal of better understanding his own work, which is both multi-disciplinary, and difficult to fully model and understand. Furthermore, collaboration among individuals from various fields can provide stimulating conversations and social connections in the future.