RECENT:Spreading, Influencing, and Cascading in Social and Information Networks
"Social consensus through the influence of committed minorities",
J. Xie, S. Sreenivasan, G. Korniss, W. Zhang, C. Lim, and B. K. Szymanski, Physical Review E 84, 011130 (2011).
News: Science News Focus, "The power of true believers", by Adrian Cho
RPI News, "Minority Rules: Scientists Discover Tipping Point for the Spread of Ideas", by Gabrielle DeMarco
Discovery News, "Minority Rules: Scientists Find the Tipping Point", by Emily Sohn
Freakonomics, "Minority Rules: Why 10 Percent is All You Need", by Matthew Philips
Communications of the ACM, "Researchers Find Tipping Point to Sway Public Opinion", by Paul Hyman
A satellite workshop at NetSci2011, Central European University, Budapest, June 7, 2011
Current graduate students: David Hunt, Pramesh
Current postdocs: Andrea Asztalos, Sameet Sreenivasan
Former graduate students: Qiming Lu, Hasan Guclu, Balazs Kozma
Current and former undergraduate students: Jefffrey Emenheiser, Adam Freese
Collaborators: Boleslaw Szymanski (Rensselaer), Zoltan Toroczkai (Notre Dame), Chjan Lim (Rensselaer)
Optimizing Synchronization, Flow, and Robustness in Complex Networks
One of the major developments of the last two decades has been the ever-increasing interconnectivity of a broad class of information networks, including physical and data network types arising in telecommunication, transportation and energy infrastructures. This interconnectivity has led to immense temporal and spatial complexity in modern networks and a critical need for basic mathematical theory and statistical modeling of complex interacting networks.
As the number of nodes in the underlying these networks increases to hundreds of thousands, fundamental questions about the stability and vulnerability of these systems (ultimately governed by the fluctuations of the dynamic processes in the respective underlying networks) must be addressed. We address precisely these issues from a fundamental network science viewpoint. In this research project, we develop and employ a unified approach and framework connecting the concepts of flow and fluctuations in network dynamics, allowing us to investigate and analyze fundamental transport, load-balancing, and synchronization problems in complex networks on a common footing. Our research shall provide tools and methods to optimize the allocation of limited network resources in order to improve the quality of service and to minimize the possibility of global network failures in both piece-time and hostile environments.
Specifically, complex biological, social, and technological systems can be often modeled by weighted networks. The network topology, together with the distribution of available link or node capacity (represented by weights) and subject to cost constraints, strongly affect the dynamics or performance of the networks. In this project, we investigate optimization in fundamental synchronization and flow problems where the weights are proportional to the degrees of the nodes connected by the edge. In the context of synchronization, these weights represent the allocation of limited resources (coupling strength), while in the associated random walk and current flow problems, they control the extent of hub avoidance, relevant in routing and search.
This research has been supported in part by DTRA, ARL-NS-CTA, and NSF-DMR (ITR)
Structural Randomness, Composite Disorder, and Electrical
Response: Resonances and Transient Delays in Complex Impedance
R. Huang, G. Korniss, and S.K. Nayak, Physical Review E 80, 045101 (Rapid Communication) (2009).
"Synchronization in Weighted Uncorrelated Complex Networks in a Noisy Environment: Optimization and Connections with Transport Efficiency'',
G. Korniss, .Physical Review E 75, 051121 (2007).
Fluctuations in Noisy Task-Completion Landscapes on Scale-Free
H. Guclu, G. Korniss, Z. Toroczkai, Chaos 17, 026104 (2007).
"Diffusion Processes on
Small-World Networks with Distance-Dependent Random Links'',
B. Kozma, M.B. Hastings, and G. Korniss, Journal of Statistical Mechanics: Theory and Experiment, P08014 (2007).
Cascading in Wireless Sensor Networks'',
Q. Lu, G. Korniss, and B.K. Szymanski, in Proceedings of the Third International Conference of Networked Sensing Systems (INSS 2006) (Transducer Research Foundation, San Diego, 2006) pp.164-171.
Small-World Resistor Networks'',
G. Korniss, M.B. Hastings, K.E. Bassler, M.J. Berryman, B. Kozma, and D. Abbott, Phyiscs Letters A 350, 324 (2006).
Landscapes in Small-World-Connected Computer Networks'',
H. Guclu, G. Korniss, M.A. Novotny, Z. Toroczkai, and Z Racz, Physical Review E 73, 066115 (2006).
Processes on Power-Law Small-World Networks'',
B. Kozma, M.B. Hastings, and G. Korniss, Physical Review Letters 95, 018701 (2005).
"Extreme Fluctuations in Small-World-Coupled Autonomous Systems
with Relaxational Dynamics",
H. Guclu and G. Korniss, Fluctuation and Noise Letters 5, L43 (2005).
Fluctuations in Small-Worlds with Relaxational Dynamics'',
H. Guclu and G. Korniss, Physycal Review E 69, 065104(R) (2004).
Roughness of Virtual Times in Parallel Discrete-Event
G. Korniss, M.A. Novotny, H. Guclu, Z. Toroczkai, and P.A. Rikvold, Science299, 677 (2003).
News: Science Perspective, "Rough Times Ahead", by Scott Kirkpatrick,
The New York Times,
Technology Research News
Scaling for Edwards-Wilkinson Relaxation in Small-World
B. Kozma, M.B. Hastings, and G. Korniss, Physical Review Letters 92, 108701 (2004).
Synchronized Computing Networks for Scalable Parallel
H. Guclu, G. Korniss, Z. Toroczkai, and M.A. Novotny, in Complex Networks, edited by E. Ben-Naim, H. Frauenfelder, and Z. Toroczkai, Lecture Notes in Physics Vol. 650 (Springer-Verlag, Berlin, 2004) 255--275.
Growth in a Small World'',
B. Kozma and G. Korniss, in Computer Simulation Studies in Condensed Matter Physics XVI, edited by D.P. Landau, S.P. Lewis, and H.-B. Schüttler, Springer Proceedings in Physics Vol. 95 (Springer-Verlag, Berlin, 2004), pp. 29-33.
Parallel Algorithms and Fluctuating Time Horizons to
Non-equilibrium Surface Growth'',
G. Korniss, Z. Toroczkai, M.A. Novotny, and P.A. Rikvold, Phys. Rev. Lett. 84, 1351 (2000).
and Social Dynamics in Large-Scale Social Networks
The current state of knowledge about the interdependency between structure, dynamics, and behaviors of large infrastructure, communication, and social networks is only at the preliminary stage. In this project, we investigate real data projected out from empirical social networks, to develop methodologies and techniques for extracting relevant information and behavioral patterns from the underlying social network. With the availability and accessibility of vast amount of data in recent years, our project will not only gain fundamental knowledge, from a network science viewpoint, but also will create methodologies and techniques that could be applied to address strategic and urgent needs aligned with national priorities for network research.
For example, fighting irregular warfare (IW) is a complex and ambiguous inherently social phenomenon: “insurgency and counterinsurgency operations focusing on the control or influence of populations, not on the control of an adversary’s forces or territory” (IW, Joint Operating Concept, 2007). To help operations in such an environment, we will develop and employ individual-based models to investigate social influencing and associated strategies in social networks. Our methods and models for community detection, community stability, and social influencing will be applicable to data sets of different types and spanning all scales, including those collected by the military.
Mainstream ideology does not freely spread into an “ideological vacuum” as standard models for social influencing assume, but instead, is severely inhibited by existing adversary or extremist opinion clusters, strengthened by the collective identity of the individuals forming these extremist clusters. Therefore we develop and employ models for social dynamics with multiple (potentially coexisting) opinions.
Investigating social dynamics and diffusion of opinions on empirical social graphs on graphs with multiple co-existing opinion clusters/communities, we will develop both heuristic and rigorous algorithmic methods to find influential nodes and to disintegrate or control adversary clusters in ideological warfare. We will systematically investigate the “influential hypothesis” in military-relevant scenarios.
Over the past few years, we and the other
network science-researchers have been shifting the focus
of network research from structure to dynamics (or
function) in real-life networks. To accelerate this
transition, in this research we study social dynamics on
large-scale empirical social networks from a fundamental
viewpoint. Our investigations will focus on the
emergence and stability of communities in social graphs
and the elimination of such communities by introducing
agents committed to stabilizing society or by exerting
This research has been supported in part by ARL NS-CTA and ONR
"Social Consensus through
the Influence of Commited Minorities",
Dynamics: Emergence of a scale-free Leadership Structure and
M. Anghel, Z. Toroczkai, K.E. Bassler, and G. Korniss, Physical Review Letters 92, 058701 (2004).
News: Technology Research News
|Spatial Ecologies under
Better understanding the processes that allow one species to invade the habitat of another will help answer fundamental questions in ecology. Just as importantly, understanding the processes of biological invasion to the point of prediction should enhance our capacity to respond effectively to invading species that impose economic problems on agriculture and threaten native biodiversity across North America.
Spatially structured biotic interactions generate demographic variation at the level of individuals; this variation can govern ecological invasion and competitive coexistence. Temporal abiotic variation can directly modulate demographic rates, exerting further effects on competitors’ dynamics. Theory for individual-based spatial processes has remained distinct from theory for population dynamics in temporally variable environments. In this project, we study the interplay between discrete, stochastic spatial interactions and temporally varying environmental forces, with a focus on competitive invasion. To advance understanding of the multi-scale ecological complexity predicted by the theory, and to identify novel predictions, we apply concepts of statistical physics.
When a superior competitor’s invades an inferior resident species in a constant environment, and introduction occurs only rarely, spatial competition will be driven by local dispersal and mortality, so that we can apply the physical theory for nucleation of spatial systems. Low introduction rates and small environments may permit only single-cluster invasion by the superior competitor. Increasing the introduction rate or system size allows multi-cluster invasive growth. In both regimes, the invasion dynamics exhibits strong spatial correlation and stochasticity. Consequently, standard deterministic approaches fail to predict the behavior of the invasion process. However, for multi-cluster invasion, the time to competitive exclusion is approximated accurately by Avrami’s law, an analytical framework of nucleation theory. The Avrami result links individual-level propagation and mortality rates to community-level features of spatial invasion, a major research objective in ecology. We are developing an integrative theory of stochastic spatial growth; judicious parameter choice should let the theory address data on invasive fronts ranging from the within-habitat to the biogeographic scale. To study invasion in two-dimensional environments, we analyze stochastic “roughening” of advancing fronts; several results apply to a broad class of spatial-growth models. Analysis has yielded novel predictions; habitat size (length of the front) should affect both invasion velocity and the expected location of the maximal invasive incursion beyond the front’s mean position.
Most importantly, we are developing a new theory for spatial competition in a time-varying environment. If each species is alternately favored, single-cluster invasion may admit competitive coexistence via stochastic resonance, indicating a matching of the (stochastic) spatial-dynamic and (periodic) environmental time scales. The multi-cluster mode may exhibit different limit cycles and long-term coexistence through large-scale clustering of each species, depending on comparison of the population dynamics’ characteristic time scale and the period-length of environmental variation.
This research develops new approaches to understanding stochastic, nonlinear effects in ecological systems and will lead to insights concerning the way individual-based ecological interactions may amplify or attenuate environmental variability.
This research has been supported in part by NSF-DEB (QEIB)
Related publications:"Spatial Competition and the Dynamics of Rarity in a Temporally Varying Environment",