The Statistics Seminar Speaker for October 14 will be Dr. Ji Zhu, professor of Statistics at the University of Michigan. Having earned his master's degree and PhD from Stanford University. Dr. Zhu's research interests include statistical learning and data mining, high dimensional data, and statistical network analysis. Visit his webpage for more information.
Title: Estimating Network Edge Probabilities by Neighborhood Smoothing
Abstract: The problem of estimating probabilities of network edges from the observed adjacency matrix has important applications to predicting missing links and network denoising. It has usually been addressed by estimating the graphon, a function that determines the matrix of edge probabilities, but is ill-defined without strong assumptions on the network structure. Here we propose a novel computationally efficient method based on neighborhood smoothing to estimate the expectation of the adjacency matrix directly, without making the strong structural assumptions graphon estimation requires.
The neighborhood smoothing method requires little tuning, has a competitive mean- squared error rate, and outperforms many benchmark methods on the task of link prediction in both simulated and real networks. This is joint work with Yuan Zhang and Elizaveta Levina.
Refreshments will be served after the seminar in 1181 Comstock Hall.