Dynamic Architecture of Networks

From Santa Fe Institute Events Wiki

CSSS Santa Fe 2010


Next meeting... when?


We want to study the statistical properties of evolving networks (networks that change the number of nodes/connections/weights). What are the invariants? What properties change over time?

For example, for dynamic food webs (which interest Jonathan Cannon), a change may result from an adaptive modification of species diet in response to circumstance or the addition and extinction of species. We will look for the structural and dynamic effects of these mechanisms, including the formation of characteristic motifs and stabilization.

For evolving gene regulatory networks, new regulatory units are created by duplications, destroyed by deletions, and the weights can change due to point mutations. (Borys Wrobel) has built an artificial life platform (right now, it uses a genetic algorithm) that can generate a large number of gene regulatory networks that have evolved to perform certain computations (for example, process signals). Borys would be interested in investigating what are the statistical (or other) properties of these networks and how they change with time. Here is a paper that describes (a version) of the model.

We are also interested in asking what is needed for a system dynamics to be preserved despite changes in network topology.

For computing the properties, we are thinking of using igraph and/or snap.