Presenter: Marina Meila (University of Washington)
Title: Geometric Learning beyond visualization: Explanations and Eigenflows
Abstract: Manifold learning algorithms are widely used as a visualization tool, either for exploration, or after scientific inference has been performed by other means. This talk aims to show that mathematically and statistically grounded geometric data analysis can support quantitative algorithms, that both recover the underlying low-dimensional structure of high-dimensional point clouds and validate inferences about these embeddings that are typically left to the scientist to sort out.
This talk will illustrate this paradigm in two directions. First, we ask if it is possible, in the case of scientific data where quantitative prior knowledge is abundant, to explain a data manifold by new coordinates,chosen from a set of scientifically meaningful functions?
Second, I will show how some popular manifold learning tools and applications can be recreated in the space of vector fields and flows on a manifold.
Central to this approach is the order 1-Laplacian of a manifold, \Delta_1, whose eigen-decomposition into gradient, harmonic, and curl, known as the Helmholtz-Hodge Decomposition, provides a basis for all vector fields on a manifold. We present an estimator for \Delta_1, and based on it we develop a variety of applications. Among them, visualization of the principal harmonic, gradient or curl flows on a manifold, smoothing and semi-supervised learning of vector fields, 1-Laplacian regularization. In topological data analysis, we describe the 1st-order analogue of spectral clustering, which amounts to prime manifold decomposition. Furthermore, from this decomposition a new algorithm for finding shortest independent loops follows. The algorithms are illustrated on a variety of real data sets.
Joint work with Yu-Chia Chen, Vlad Murad, Samson Koelle, Hanyu Zhang and Ioannis Kevrekidis