Statistical shape analysis via Topological data analysis

Speaker: Assistant Professor Chul Moon

As modern data applications become complex in size and structure, identifying the underlying shape and structure has become of fundamental importance. The classical approaches such as dimension reduction are challenging for handling these applications. Topological data analysis (TDA) is a rapidly developing collection of methods that focuses on the “shape” of data. TDA can uncover the underlying low-dimensional geometric and topological structures from high-dimensional datasets. TDA has been successfully applied to various areas, including biology, network data, material science, and geology, in recent years. The goal of the lecture is to introduce novel TDA methods that can capture geometric or topological information of data and make statistical inferences. Covered by Assistant Professor Chul Moon, this lecture aims to familiarize these new methods along with their applications to various types of data.