Abstract: Discrete events are a sequence of observations consisting of event time, location, and possibly “marks” with additional event information. Such event data is ubiquitous in modern applications, such as social networks, seismic activities, police reports, neuronal spike trains, and disease spread counts. We are particularly interested in capturing the complex dependence of the discrete events data, particularly estimating how nodes interact with each other, such as the triggering or inhibiting effects of the historical events on future events. This helps us recovery network topology, perform causal inference, understand spatio-temporal dynamics, and make predictions. Motivated by popular Hawkes processes, we introduce a new general modeling approach for capturing spatio-temporal interaction, which enjoys computationally efficient model estimation procedures. We establish statistical guarantees by connecting to a modern convex optimization theory of solving variational inequality. The good performance of the proposed method is illustrated using several real-world data sets.
Biography: Yao Xie is an Associate Professor and Harold R. and Mary Anne Nash Early Career Professor at Georgia Institute of Technology in the H. Milton Stewart School of Industrial and Systems Engineering, and an Associate Director of the Machine Learning Center. She received her Ph.D. in Electrical Engineering (minor in Mathematics) from Stanford University and was a Research Scientist at Duke University. Her research areas are statistics (particularly change-point detection and spatio-temporal data modeling), machine learning, and signal processing. She received the National Science Foundation (NSF) CAREER Award in 2017. She is currently an Associate Editor for IEEE Transactions on Signal Processing, Sequential Analysis: Design Methods and Applications, INFORMS Journal on Data Science, and serves on the Editorial Board of Journal of Machine Learning Research.