Abstract: Topological data analysis (TDA) is a mathematical method for analyzing the shape and structure of complex data sets. It has recently been gaining popularity in the machine learning community, due to its unique ability to uncover hidden patterns and features that are not easily identifiable through traditional methods.
In this talk, we will provide a comprehensive introduction to the basics of TDA, including key concepts such as topological spaces, homeomorphisms, and persistent homology. We will then delve into the details of how TDA can be applied in the context of machine learning, including the use of tools such as the Mapper algorithm and the TDA package in R.
Furthermore, we will discuss the advantages of using TDA in machine learning, such as its ability to handle high-dimensional data, its robustness to noise and missing data, and its interpretability. We will also present several real-world examples and case studies that illustrate the power of TDA in uncovering insights from complex data sets.
Bio: Christian is Machine Learning Technical Leader at Mercado Libre, the largest e-commerce/fintech company in Latin America, where he dedicates his efforts to creating tools for monitoring and quality of learning models. He is a Computer Engineer and Master in Science with a major in Astronomy from UNAM (Universidad Nacional Autonoma de Mexico). He is a "Xoogler" and has more than 15 years of experience in the field of machine learning. He has lectured in almost a dozen countries.