{"id":"arxiv/2604.04272","url":"https://arxiv.org/abs/2604.04272","tags":["math.ST"],"score":0,"title":"Theoretical Foundations of Principal Manifold Estimation with Non-Euclidean Templates","domain":"math.ST","source":"arxiv","authors":["Kun Meng","Christopher Perez"],"reviews":0,"subject":"math.ST","summary":"We develop a rigorous theoretical framework for principal manifold estimation that recovers a latent low-dimensional manifold from a point cloud observed in a high-dimensional ambient space. Our framework accommodates manifolds with general, potentially non-Euclidean topology, which can be inferred using tools from topological data analysis. Using the theory of Sobolev spaces on Riemannian manifolds, we establish that the proposed principal manifolds are well defined, prove convergence of the iterative algorithm used to compute them, and show consistency of the finite-sample estimator.…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-04-05","reactions":{}}