product description page
Algorithmic Advances in Riemannian Geometry and Applications : For Machine Learning, Computer Vision,
About this item
This volume presents a comprehensive treatment of Riemannian geometry as a mathematical and computational framework for many problems in machine learning, statistics, optimization, and computer vision. The chapters in the volume are written by leading experts in the field and showcase the latest advances made recently, both theoretically and algorithmically. Examples include the geometrical foundation of Hamiltionian Monte Carlo, large-scale Riemannian optimization of low-rank matrices for matrix completion, and kernel methods on symmetric positive definite matrices for visual object recognition.