What is SPINet?

A vast array of systems form complex networks, such as transport infrastructures, social interactions, communications networks, protein networks, and the Internet. There is compelling evidence that both the network evolution and the dynamics on the network, such as transport of material or information, obey a small set of universal principles. Finding these principles is paramount for the development of tools and methods which will allow the prediction of the dynamics on different kinds of networks and the optimization of specific properties, such as robustness to breakdown, transport, dissipation, and cost. A key problem in finding these principles will be the development of the proper analytical and computational methods to accurately characterize the networks and their dynamics. Since these systems can consist of large 105 - 1010 collections of individual elements, any practical tools and principles have to be statistical in nature. In addition, complex networks are neither completely regular nor purely random structures, a property rooted  in the typically stochastic process that leads to their formation and evolution. These properties make these systems ideal for the Statistical Physics approach.

SPINet is a team of people at Los Alamos Laboratory and collaborators from Notre-Dame, Rensselaer, Boston, Northwestern and Rome, assembled to develop a consistent theory for the Dynamics of Complex Networks with Infrastructure Applications. The techniques for analysis, characterization and optimization for this class of systems will also be highly valuable for a wide range of disciplines in the basic sciences in which the nature and dynamics of networks plays a crucial role, such as Biological and Social Networks.

"Networks: Structure, Dynamics and Function": 23-rd CNLS Annual Conference
  the largest conference yet on Complex Networks.  Organized by the Center for Nonlinear Studies (CNLS) and LANL.
  Organizers: Z. Toroczkai, E. Ben-Naim, H. Frauenfelder, P. Swart, B. McMahon, G. Istrate, Y. Jiang, S. Eubank, P. Fenimore and C. Reichhardt

book, Lecture Notes in Physics Series:  "Complex Networks" eds. E. Ben-Naim, H. Frauenfelder and Z. Toroczkai, 2004.

The SPINet Team:

Theoretical Division, Complex Systems Group

Computer and Computational Sciences Division:

Decision Applications Division:

CNLS

 External Collaborators:  

Previous LANL project involvements by team members:
(partial list)

 EpiSims The project goal is to work out vaccination and mitigation strategies for smallpox and similar epidemics for large-scale urban populations. The project requires the analysis of massive people contact-networks from the data. The contact graph data are generated by a large-scale virtual city simulation instantiated by demographic land-use census data and people diaries. Team collaborators: S. Eubank, M. Marathe, G. Istrate, Z. Toroczkai and E. Ben-Naim.

 AdHopNet The project focuses on the development of ad-hoc and hybrid mobile communication networks, in particular on designing efficient routing protocols for packet transmission. In contrast with the current centralized approach based on  fixed base-station layouts, the ad-hoc communications use the instantaneous configuration of the mobile devices to perform the routing. Team collaborators: M. Marathe, G. Istrate and Z. Toroczkai.

TAP The Terrorist Anticipation Project, sponsored by DTRA, aims at creating an agent-based, dynamic model for social unrest, and anticipating the likelihood of a terrorist attack as a result of the dynamic interplay between the cooperation on the social network level among the agents, availability of resources, and exogenous pressure factors (such as economical embargo's and military actions). Team collaborators: Ed MacKerrow, David H. Sharp and Z. Toroczkai.

Vortex Cellular Automata and Logic Devices With the advances in nanoscale fabrication and control, new kinds of nanoscale devices can now be considered. One promising example has been quantum dot cellular automata where the locations of just one or two charges can be used to represent 0 or 1 for the basic building units for logic devices. We have been considering a similar approach with superconducting vortices in nanostructured arrays to create vortex cellular automata. We have proposed particular dot arrangements to form NAND and AND gates as well as wire crossing geometries. In addition to new types of logic circuits, we have also been considering new types of mechanical devices such as gears. In addition to the technological applications, there are also many scientific issues that we are exploring in these systems such as using various types of deterministic ratchets for soliton propagation. Team collaborators: C. Reichhardt and M. Hastings.

Colloids Colloidal crystals are an ideal system in which to study the general problem of ordering and dynamics in 2D, since the particle size permits direct imaging of the particle locations and motion. A considerable amount of work has been conducted on the melting of 2D colloidal crystals in the absence of a substrate. In addition, a number of experimental and theoretical studies have considered colloidal crystallization and melting in 2D systems with periodic 1D and 2D substrates, where a rich variety of crystalline states can be stabilized. Colloid crystals are also ideal for studying the ordering and dynamics of an elastic media interacting with random substrates, a problem that is relevant to a wide variety of systems such as superconducting vortices, Wigner crystals, and charge-density waves. Open issues include the nature of the dynamical response to applied forces, as well as whether an order to disorder transition occurs as the strength of the random substrate increases. Team collaborators: C. Reichhardt and M. Hastings.

Agent-Collective Optimization through Influence Network Design This project focuses on investigating the effects of communication across a social network by the agents of a multi-agent game. We have shown that when the multi-agent game is a competition game, the social network generates a dynamic influence sub-network, which governs the game's evolution. We have shown that by tuning the connectivity properties of the social network, the collective's behavior can dramatically be improved and the game made efficient. We apply these results to a simplistic model of a market. Team collaborators: Z. Toroczkai and M. Anghel.

Nonlinear Symbolic Time Series Analysis Symbolic nonlinear time series analysis methods have the potential for analyzing nonlinear data efficiently with low sensitivity to noise. In symbolic nonlinear time series analysis a time series for a fixed delay is partitioned into a small number (called the alphabet size) of cells labeled by symbols, creating a symbolic time series. Symbolic methods involve computing the statistics of words made from the symbolic time series. We have shown methods of computing the metric entropy for nonlinear flows as well as for nonlinear maps. We also proposed a method of computing the information dimension appropriate to symbolic analysis. Team collaborators: Z. Toroczkai and M. Anghel.

Granular Media Granular media is unique as it involve macroscopically sized particles and large energy dissipation. Yet, they can exhibit solid-like or liquid-like behavior. They also exhibit interesting collective phenomena such size segregation, pattern formation, shock waves, slow density relaxation, and density inhomogeneities. We investigated experimentally the distribution of configurations of a ring with an elementary topological constraint, a figure-8 twist. Using a system far from thermal equilibrium, a vibrated granular chain, we have shown  that configurations where one loop is small and the second is large are strongly preferred. Despite the highly non-equilibrium nature of the system, our results are consistent with recent predictions for equilibrium properties of topologically-constrained polymers. Team collaborators: E. Ben-Naim and M. Hastings.