Fast Manifold Learning Based on Riemannian Normal Coordinates

  title={Fast Manifold Learning Based on Riemannian Normal Coordinates},
  author={Anders Brun and Carl-Fredrik Westin and Magnus Herberthson and Hans Knutsson},
We present a novel method for manifold learning, i.e. identi fication of the low-dimensional manifold-like structure present in a s et of data points in a possibly high-dimensional space. The main idea is derived f rom the concept of Riemannian normal coordinates. This coordinate system is i n a way a generalization of Cartesian coordinates in Euclidean space. We tra nslate this idea to a cloud of data points in order to perform dimension reduction . Our implementation currently uses Dijkstra’s… CONTINUE READING
Highly Cited
This paper has 29 citations. REVIEW CITATIONS