Statistical embedding: beyond principal components
From MaRDI portal
\(t\)-SNEdiffusion mappinggraph spectral theoryISOMAPlocal linear methodMappermultidimensional scalingneighborhood sampling strategiesnetwork embeddingnon-linear principal componentpersistence diagrampersistent homologyprincipal componentrandom projectionreproducing kernel Hilbert spaceskip-Gramspectral embeddingstatistical embeddingstochastic block modelingtopological data analysis and embeddingvisualization
Abstract: There has been an intense recent activity in embedding of very high dimensional and nonlinear data structures, much of it in the data science and machine learning literature. We survey this activity in four parts. In the first part we cover nonlinear methods such as principal curves, multidimensional scaling, local linear methods, ISOMAP, graph based methods and diffusion mapping, kernel based methods and random projections. The second part is concerned with topological embedding methods, in particular mapping topological properties into persistence diagrams and the Mapper algorithm. Another type of data sets with a tremendous growth is very high-dimensional network data. The task considered in part three is how to embed such data in a vector space of moderate dimension to make the data amenable to traditional techniques such as cluster and classification techniques. Arguably this is the part where the contrast between algorithmic machine learning methods and statistical modeling, the so-called stochastic block modeling, is at its greatest. In the paper, we discuss the pros and cons for the two approaches. The final part of the survey deals with embedding in , i.e. visualization. Three methods are presented: -SNE, UMAP and LargeVis based on methods in parts one, two and three, respectively. The methods are illustrated and compared on two simulated data sets; one consisting of a triplet of noisy Ranunculoid curves, and one consisting of networks of increasing complexity generated with stochastic block models and with two types of nodes.
Cites work
- A nonparametric view of network models and Newman–Girvan and other modularities
- An information criterion for variable selection in support vector machines
- Asymptotic normality of maximum likelihood and its variational approximation for stochastic blockmodels
- Asymptotic theory for density ridges
- Augmented Implicitly Restarted Lanczos Bidiagonalization Methods
- Co-clustering directed graphs to discover asymmetries and directional communities
- Community structure in social and biological networks
- Computing persistent homology
- Consistency of spectral clustering in stochastic block models
- Detection of Abnormal Behavior Via Nonparametric Estimation of the Support
- Diffusion maps
- Dynamic stochastic block models: parameter estimation and detection of changes in community structure
- Empirical Analysis of an Evolving Social Network
- Extensions of Lipschitz mappings into a Hilbert space
- Extremal properties of principal curves in the plane
- Fast unfolding of communities in large networks
- Finding the homology of submanifolds with high confidence from random samples
- Grouped network vector autoregression
- Homological algebra and data
- How to Draw a Graph
- scientific article; zbMATH DE number 6378173 (Why is no real title available?)
- scientific article; zbMATH DE number 3149985 (Why is no real title available?)
- scientific article; zbMATH DE number 7646025 (Why is no real title available?)
- Hypothesis testing for automated community detection in networks
- Information criteria and statistical modeling.
- Laplacian Eigenmaps for Dimensionality Reduction and Data Representation
- Latent Space Approaches to Social Network Analysis
- Likelihood-based model selection for stochastic block models
- Locally defined principal curves and surfaces
- Manifold estimation and singular deconvolution under Hausdorff loss
- Minimax rates for estimating the dimension of a manifold
- Modularity based community detection in heterogeneous networks
- Multidimensional scaling. I: Theory and method
- Network vector autoregression
- Networks
- Nonlinear Estimators and Tail Bounds for Dimension Reduction in l 1 Using Cauchy Random Projections
- Nonparametric ridge estimation
- Pairwise local Fisher and naive Bayes: improving two standard discriminants
- Predicting clinical outcomes in glioblastoma: an application of topological and functional data analysis
- Principal component analysis.
- Principal Curves
- Random-projection ensemble classification. (With discussion).
- Reducing the Dimensionality of Data with Neural Networks
- RELATIONS BETWEEN TWO SETS OF VARIATES
- Representation learning for dynamic graphs: a survey
- Selecting the number of components in principal component analysis using cross-validation approximations
- Self-organized formation of topologically correct feature maps
- Some recent trends in embeddings of time series and dynamic networks
- Spectral clustering and the high-dimensional stochastic blockmodel
- Statistical analysis and parameter selection for Mapper
- Statistical dependence: beyond Pearson's
- Statistical modeling using local Gaussian approximation
- Theoretical foundations of the potential function method in pattern recognition learning
- Topological data analysis of single-cell Hi-C contact maps
- Topological persistence and simplification
- Topology and data
- Visualizing data using t-SNE
This page was built for publication: Statistical embedding: beyond principal components
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6181742)
