A Discussion on Dimensionality Reduction and Data Science Education
Dr. Reinaldo Sanchez Arias
St. Thomas University, Miami
Abstract: The field of machine learning deals with the development and application of computer systems that can learn from data, where learning has taken place when a metric for performance on a given task improves after exposure to data. Computers running a variety of machine learning algorithms can recognize and cluster specific objects in pictures, discriminate subtle differences and trends in complex data, predict the class of a sample among a set of possible categories, or recommend a following step based on historical data and association rules. In this talk we discuss a framework for dimensionality reduction based on machine learning approaches that can be used for reduced-order modeling, with the purpose of representing the dominant behavior of the system of interest. Computational tools for data analytics, predictive modeling, and the general data science workflow for multidisplicinary research will also be discussed.
Bio: Reinaldo (Rei) Sanchez-Arias earned a Bachelor of Science degree in Mathematics from Universidad del Valle in Cali, Colombia. In Fall 2008 he started his doctoral studies at The University of Texas at El Paso (UTEP). During his years at UTEP he was involved in research projects for the Army High Performance Computing Research Center (AHPCRC). He obtained a Ph.D. in Computational Science from UTEP in 2013, and completed a postdoctoral researcher appointment for the AHPCRC before joining the Applied Mathematics Department at Wentworth Institute of Technology in Boston. In 2016, he joined St Thomas University in Miami, and serves as the Program Director for the MS in Big Data Analytics. His latest research efforts have been focused on the design of fast algorithms for sparse representation and dimensionality reduction, and data science education at all levels.
Read More