Geometry of covariance matrices, covariance operators, and Gaussian processes – theory and applications
Quang Minh Ha received his BSc degree in Mathematics and Computer Science from Monash University, Australia, and PhD degree in mathematics from Brown University, USA, under the supervision of Steve Smale. He is currently a Unit Leader at the RIKEN Center for Advanced Intelligence Project in Tokyo, Japan. Prior to joining RIKEN, he was a researcher at the Istituto Italiano di Tecnologia in Genoa, Italy. He has also held positions at the University of Vienna, Austria, and the Humboldt University of Berlin, Germany. His current research interests focus on the application of functional analysis and differential geometry in the formulation of theories and algorithms in machine learning, statistics, computer vision, image and signal processing.
Covariance matrices, covariance operators, Gaussian measures, and Gaussian processes play important roles in many areas of machine learning, statistics, and numerous application domains. Of particular importance in the study of these objects and their applications are the similarity functions between them. In this talk, we present an overview of some recent generalizations of non-Euclidean distances between covariance matrices and Gaussian measures on Euclidean space to the infinite-dimensional setting of covariance operators and Gaussian processes. Our focus is on the generalization of the Fisher-Rao distance and related quantities (from Information Geometry) and the Entropic Regularization of the 2-Wasserstein distance (from Optimal Transport). Our formulations all admit closed form expressions that can be readily applied in practice, particularly in the reproducing kernel Hilbert space (RKHS) setting. The mathematical formulations will be accompanied by numerical experiments in computer vision.
Related seminars
Stony Brook University
Washington State University
University of Melbourne
National University of Singapore