A Convex Framework for Contagion Control in Networks

With networks becoming pervasive, the last decade saw an increase in research on dynamics of spreading processes in networks. While a number of well understood epidemic spreading models have been developed, little to no attention has been paid to epidemic control strategies; beyond heuristics usually based on network centrality measures and limited to undirected networks. Since epidemic control resources are typically limited, the problem of optimally allocating resources to control an outbreak is of interest.

In the talk, I will consider the well-known Susceptible-Infected-Susceptible spreading model and present the problem of minimum cost resource allocation to control an epidemic outbreak in a networked population. Via a convex optimization framework, I would present optimal solutions to the epidemic control problem on generic weighted, directed networks comprising non-identical nodes. Based on our spreading model, we express the problem of controlling an epidemic outbreak in terms of spectral conditions involving the Perron-Frobenius eigenvalue. This enables formulation of the epidemic control problem as a Geometric Program, for which we derive a convex characterization guaranteeing optimality of solutions. Finally, I will present a fully distributed solution, validations of our results on data and present some new results on traffic shaping in the presence of a contagion.

Speaker Bio: Chinwendu Enyioha is a post doctoral researcher in the Electrical and Systems Engineering department at the University of Pennsylvania, where he completed his Ph.D in 2014 with a focus in network science, systems, optimization and control theory. Prior to that, he received the B.Sc in Mathematics from Gardner-Webb University (GWU) Summa cum laude in 2008, and the M.Sc in Electrical Engineering from University of Pennsylvania in 2010. He is affiliated with the GRASP Laboratory and PRECISE Center. His current research interests include control of networks, distributed computation and optimization over networks with applications to cyber-physical systems. Chinwendu Enyioha was named a Fellow of the Ford Foundation by the National Research Council; and a William Fontaine Scholar at the University of Pennsylvania. He has also received the Mathematical Association of America (MAA) Walt and Susan Patterson Award.

Note: Visitors from outside NIST must contact Cathy Graham; (301) 975-3800; at least 24 hours in advance.