Dimension estimation using random connection models
From MaRDI portal
Publication:4637068
Abstract: Information about intrinsic dimension is crucial to perform dimensionality reduction, compress information, design efficient algorithms, and do statistical adaptation. In this paper we propose an estimator for the intrinsic dimension of a data set. The estimator is based on binary neighbourhood information about the observations in the form of two adjacency matrices, and does not require any explicit distance information. The underlying graph is modelled according to a subset of a specific random connection model, sometimes referred to as the Poisson blob model. Computationally the estimator scales like n log n, and we specify its asymptotic distribution and rate of convergence. A simulation study on both real and simulated data shows that our approach compares favourably with some competing methods from the literature, including approaches that rely on distance information.
Recommendations
- Intrinsic dimension identification via graph-theoretic methods
- Local intrinsic dimension estimation by generalized linear modeling
- Novel high intrinsic dimensionality estimators
- Intrinsic dimension estimation based on local adjacency information
- Intrinsic dimension estimation of manifolds by incising balls
Cites work
- A class of Rényi information estimators for multidimensional densities
- A new look at the statistical model identification
- An Algorithm for Finding Intrinsic Dimensionality of Data
- An Intrinsic Dimensionality Estimator from Near-Neighbor Information
- Continuum Percolation
- Dimension reduction techniques and the classification of bent double galaxies
- Dimension reduction: a guided tour
- Empirical geometry of multivariate data: a deconvolution approach.
- Empirical graph Laplacian approximation of Laplace–Beltrami operators: Large sample results
- Estimating the dimension of a model
- Extreme-value-theoretic estimation of local intrinsic dimensionality
- Finding nearest neighbors in growth-restricted metrics
- Hessian eigenmaps: Locally linear embedding techniques for high-dimensional data
- High-dimensional classification using features annealed independence rules
- scientific article; zbMATH DE number 3150484 (Why is no real title available?)
- scientific article; zbMATH DE number 3864315 (Why is no real title available?)
- scientific article; zbMATH DE number 847265 (Why is no real title available?)
- Measuring the strangeness of strange attractors
- Minimax rates for estimating the dimension of a manifold
- Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis
- Nonlinear Dimensionality Reduction
- Nonmetric multidimensional scaling. A numerical method
- On a continuum percolation model
- On Local Intrinsic Dimension Estimation and Its Applications
- Percolation
- Some theory for Fisher's linear discriminant function, `naive Bayes', and some alternatives when there are many more variables than observations
- The analysis of proximities: Multidimensional scaling with an unknown distance function: I, II
- The intrinsic dimensionality of signal collections
- The self-organizing map
Cited in
(4)- Intrinsic dimension estimation based on local adjacency information
- Estimation of local degree distributions via local weighted averaging and Monte Carlo cross-validation
- scientific article; zbMATH DE number 5713166 (Why is no real title available?)
- Intrinsic dimension identification via graph-theoretic methods
This page was built for publication: Dimension estimation using random connection models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4637068)
