Understanding networks is the key to understanding interaction. Today, networks are acquiring an increasing importance both in design of modern engineering systems (e.g. the power grid, and BigData sensing, communication, and parallel-processing, multicore processors), and also in understanding natural systems (e.g. social and biological networks, the Human Connectome Project, etc.). Yet, theories of networks in different fields -- namely, communication, computation, and biological/social -- have developed with little cross-pollination. The goal of this class is to equip researchers interested in any of these fields with technical tools of information-flow analysis, and background concepts on networks, helping them to collaborate beyond their current area of interest, as well as identify potential unsolved problems for future research. Using benefit of hindsight, this special-topics graduate course will bring together the understanding of information flows in networks across the fields of communication, computation, and neuroscience. The first half of this class will focus on mathematical techniques of analyzing flows of information in communication and computation networks. We will answer questions of the nature: what is a provably efficient network structure? In the second half of the course, we will study the application of these techniques both in design of engineering networks, and in understanding biological ones, with emphasis on neuronal networks.
Number of Units: 12
Pre-requisites: undergraduate probability, linear algebra