Proposal Description

Recent advances in computing and measurement technologies have led to an explosion in the amount of data that are being collected in all areas of application. Much of these data have complex structure, in the form of text, images, video, audio, streaming data, and so on. The proposed session focuses on one important class of problems, viz, data with network structures. Such data are common in diverse engineering and scientific areas, such as biology, computer science, economics, business, epidemiology, sociology and so on. As a consequence, research on networks has steadily increased in recent years, and has also appeared in leading science publications. For example, Nature has published several reviews on the subject, Science and PNAS devoted special issues to it. While there has been extensive research on networks and practical successes, much of it happen outside the field of Statistics, and our theoretical and methodological understanding of their statistical properties is still limited. This offers statisticians numerous open questions and opportunities to be involved and allows statisticians to play critical roles. The scope of the invited talks in the proposed session ranges from characterizing and modeling network structures based on statistical principles, to exploiting the network structure as additional information to develop statistical machine learning methods. The five speakers (including three males and two females) in the proposed session are a mixture of outstanding early- and mid- career statisticians, who come from different geographic locations (three are from the U.S., one from Europe, and one from Asia) and have extensive experience in computing, theory, methodology and applications. It is my belief that in order for the statistics community to grow healthily, it is vital to provide opportunities for the early- and mid- career statisticians to disseminate their research findings. The five invited speakers will present their most recent progresses in the area of statistical network analysis and provide new directions from the statistical perspective. Their unique insights will be valuable to the broad scientific community working on cutting-edge network problems.