Data driven estimation of Laplace-Beltrami operator
Frederic Chazal · Ilaria Giulini · Bertrand Michel
Keywords:
Spectral Methods
(Other) Statistics
Graph-based Learning
Nonlinear Dimension Reduction and Manifold Learning
2016 Poster
Abstract
Approximations of Laplace-Beltrami operators on manifolds through graph Laplacians have become popular tools in data analysis and machine learning. These discretized operators usually depend on bandwidth parameters whose tuning remains a theoretical and practical problem. In this paper, we address this problem for the unormalized graph Laplacian by establishing an oracle inequality that opens the door to a well-founded data-driven procedure for the bandwidth selection. Our approach relies on recent results by Lacour and Massart (2015) on the so-called Lepski's method.
Chat is not available.
Successful Page Load