Carnegie Mellon University
file.pdf (1.92 MB)

Multi-Manifold Semi-Supervised Learning

Download (1.92 MB)
journal contribution
posted on 2009-04-01, 00:00 authored by Andrew B. Goldberg, Xiaojin Zhu, Aarti Singh, Zhiting Xu, Robert Nowak

We study semi-supervised learning when the data consists of multiple intersecting manifolds. We give a finite sample analysis to quantify the potential gain of using unlabeled data in this multi-manifold setting. We then propose a semi-supervised learning algorithm that separates different manifolds into decision sets, and performs supervised learning within each set. Our algorithm involves a novel application of Hellinger distance and size-constrained spectral clustering. Experiments demonstrate the benefit of our multi-manifold semi-supervised learning approach


Publisher Statement

Copyright 2009 by the authors



Usage metrics