Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Learn about the challenges and solutions in principal component analysis through this mathematics research talk that addresses the 'curse of isotropy' - a critical issue affecting the interpretation of principal components with similar eigenvalues. Explore how rotational variability can impact analysis results and discover a novel approach using probabilistic covariance models parameterized with flags of subspaces. Master the concept of principal subspace analysis as a more reliable alternative to traditional principal components, with demonstrations of its practical applications across various scientific datasets. The presentation, delivered at the Erwin Schrödinger International Institute's Thematic Programme on "Infinite-dimensional Geometry," offers a straightforward methodology that promises improved interpretability in exploratory data analysis.
