Learning a metric space for neighborhood topology estimation: Applications to manifold learning
Full Text: Moustafa13.pdf
Manifold learning algorithms rely on a neighbourhood graph to provide an estimate of the data’s local topology. Unfortunately, current methods for estimating local topology assume local Euclidean geometry and locally uniform data density, which often leads to poor data embeddings. We address these shortcomings by proposing a framework that combines local learning with parametric density estimation for local topology estimation. Given a data set mathcalD ⊂mathcalX, we first estimate a new metric space (mathbbX,d_mathbbX) that characterizes the varying sample density of mathcalX in mathbbX, then use (mathbbX,d_mathbbX) as a new (pilot) input space for the graph construction step of the manifold learning process. The proposed framework results in significantly improved embeddings, which we demonstrated objectively by assessing clustering accuracy.
Citation
K. Abou-Moustafa, D. Schuurmans, F. Ferrie. "Learning a metric space for neighborhood topology estimation: Applications to manifold learning". Asian Conference on Machine Learning, (ed: Cheng Soon Ong, Tu Bao Ho), pp 341-356, November 2013.| Keywords: | |
| Category: | In Conference |
| Web Links: | PMLR |
BibTeX
@incollection{Abou-Moustafa+al:ACML13,
author = {Karim T. Abou-Moustafa and Dale Schuurmans and Frank P. Ferrie},
title = {Learning a metric space for neighborhood topology estimation:
Applications to manifold learning},
Editor = {Cheng Soon Ong, Tu Bao Ho},
Pages = {341-356},
booktitle = {Asian Conference on Machine Learning},
year = 2013,
}Last Updated: February 19, 2020Submitted by Sabina P
