Colloquium: "Big Data: A Perspective from Statistical Foundation of Data Science"

Abstract: The rapid development of information technology makes it possible to collect massive amounts of multidimensional, multimodal data with high dimensionality in diverse fields of science and engineering. New statistical and machine learning methods have been developing continuously to solve challenging problems arising out of these complex systems. For example, we consider a large neuroimaging classification problem with complex dependence structures. Various machine learning techniques were found useful for analyzing such complex data. However, their statistical and computational foundations are not well established. This talk will discuss several learning procedures and associated computational, theoretical, and practical challenges.

Host: Jose Figueroa-Lopez