A nonlinear dimensionality reduction framework using smooth geodesics. (March 2019)

• Record Type: Journal Article
• Authors: Gajamannage, Kelum; Paffenroth, Randy; Bollt, Erik M.
• Is Part Of: Pattern recognition. Volume 87(2019:Mar.)
• Page Start: 226
• Page End: 236
• Publication Date: 2019-03
• Subjects: Manifold -- Nonlinear dimensionality reduction -- Smoothing spline -- Geodesics -- Noisy measurements; 57M50 -- 57N16 -- 58D15; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2018.10.020
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: Our algorithm reveals underlying smooth manifolds of high-dimensional data. The smoothing spline approach ensures the smoothness of a corrupted manifold. The algorithm handles noise and sparsity gracefully. Performance versus neighborhood size, smoothness, sparsity, and noise are analyzed. Compared to Isomap, embeddings of face images and hand written digits are improved. Abstract: Existing dimensionality reduction methods are adept at revealing hidden underlying manifolds arising from high-dimensional data and thereby producing a low-dimensional representation. However, the smoothness of the manifolds produced by classic techniques over sparse and noisy data is not guaranteed. In fact, the embedding generated using such data may distort the geometry of the manifold and thereby produce an unfaithful embedding. Herein, we propose a framework for nonlinear dimensionality reduction that generates a manifold in terms of smooth geodesics that is designed to treat problems in which manifold measurements are either sparse or corrupted by noise. Our method generates a network structure for given high-dimensional data using a nearest neighbors search and then produces piecewise linear shortest paths that are defined as geodesics. Then, we fit points in each geodesic by a smoothing spline to emphasize the smoothness. The robustness of this approach for sparse and noisy datasets is demonstrated by the implementation of the method on synthetic and real-world datasets.