Data spectroscopy: eigenspaces of convolution operators and clustering
From MaRDI portal
Abstract: This paper focuses on obtaining clustering information about a distribution from its i.i.d. samples. We develop theoretical results to understand and use clustering information contained in the eigenvectors of data adjacency matrices based on a radial kernel function with a sufficiently fast tail decay. In particular, we provide population analyses to gain insights into which eigenvectors should be used and when the clustering information for the distribution can be recovered from the sample. We learn that a fixed number of top eigenvectors might at the same time contain redundant clustering information and miss relevant clustering information. We use this insight to design the data spectroscopic clustering (DaSpec) algorithm that utilizes properly selected eigenvectors to determine the number of clusters automatically and to group the data accordingly. Our findings extend the intuitions underlying existing spectral techniques such as spectral clustering and Kernel Principal Components Analysis, and provide new understanding into their usability and modes of failure. Simulation studies and experiments on real-world data are conducted to show the potential of our algorithm. In particular, DaSpec is found to handle unbalanced groups and recover clusters of different shapes better than the competing methods.
Recommendations
Cites work
- scientific article; zbMATH DE number 3671573 (Why is no real title available?)
- scientific article; zbMATH DE number 1304258 (Why is no real title available?)
- scientific article; zbMATH DE number 1950576 (Why is no real title available?)
- scientific article; zbMATH DE number 823069 (Why is no real title available?)
- Consistency of spectral clustering
- Contour and texture analysis for image segmentation
- Horseshoes in multidimensional scaling and local kernel methods
- Random matrix approximation of spectra of integral operators
Cited in
(18)- On the approximation of kernel functions
- Covariate regularized community detection in sparse graphs
- Identifiability of nonparametric mixture models and Bayes optimal clustering
- Encoding dissimilarity data for statistical model building
- On the predictive potential of kernel principal components
- Learning low-dimensional nonlinear structures from high-dimensional noisy data: an integral operator approach
- The geometry of kernelized spectral clustering
- Clustering based on eigenvectors of the adjacency matrix
- Robust classification of multivariate time series by imprecise hidden Markov models
- Spectral Clustering, Bayesian Spanning Forest, and Forest Process
- Spectral analysis of weighted Laplacians arising in data clustering
- A fast contour-integral eigensolver for non-Hermitian matrices
- Clustering risk in nonparametric hidden Markov and i.i.d. models
- Range shifts under constant-speed and accelerated climate warming
- scientific article; zbMATH DE number 7626762 (Why is no real title available?)
- Learning by unsupervised nonlinear diffusion
- A spectral series approach to high-dimensional nonparametric regression
- Eigen-analysis of nonlinear PCA with polynomial kernels
This page was built for publication: Data spectroscopy: eigenspaces of convolution operators and clustering
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1043719)
