Diffusion maps for changing data

DOI10.1016/j.acha.2013.03.001zbMath1302.62122arXiv1209.0245MaRDI QIDQ2252180

Matthew J. Hirn, Ronald R. Coifman

Publication date: 16 July 2014

Published in: Applied and Computational Harmonic Analysis (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1209.0245

zbMATH Keywords

kernel method; dimensionality reduction; spectral graph theory; manifold learning; graph Laplacian; dynamic graphs; diffusion distance

Mathematics Subject Classification ID

62H11: Directional data; spatial statistics

62G10: Nonparametric hypothesis testing

62J99: Linear inference, regression

68U10: Computing methodologies for image processing

60J20: Applications of Markov chains and discrete-time Markov processes on general state spaces (social mobility, learning theory, industrial processes, etc.)

05C50: Graphs and linear algebra (matrices, eigenvalues, etc.)

94A08: Image processing (compression, reconstruction, etc.) in information and communication theory

Related Items

Recovering Hidden Components in Multimodal Data with Composite Diffusion Operators, Data Based Construction of Kernels for Classification, Introducing User-Prescribed Constraints in Markov Chains for Nonlinear Dimensionality Reduction, The pre-image problem for Laplacian Eigenmaps utilizing L ₁ regularization with applications to data fusion, Sparse control of alignment models in high dimension, Local kernels and the geometric structure of data, Spatial-spectral operator theoretic methods for hyperspectral image classification, Time coupled diffusion maps, A unified framework for harmonic analysis of functions on directed graphs and changing data, Determining baseline profile by diffusion maps, Diffusion maps-aided neural networks for the solution of parametrized PDEs, Diffusion representation for asymmetric kernels, Multi-view kernel consensus for data analysis, The stability of the first Neumann Laplacian eigenfunction under domain deformations and applications, Manifold learning for parameter reduction, Diffusion maps tailored to arbitrary non-degenerate Itô processes, Diffusion nets, Partitions of networks that are robust to vertex permutation dynamics

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• A spectral notion of Gromov-Wasserstein distance and related methods
• Traceable integral kernels on countably generated measure spaces
• Linear-projection diffusion on smooth Euclidean submanifolds
• Bi-stochastic kernels via asymmetric affinity functions
• Diffusion maps
• Vector diffusion maps and the connection Laplacian
• Kernels of Trace Class Operators
• Laplacian Eigenmaps for Dimensionality Reduction and Data Representation
• Ergodic theory and visualization. I. Mesochronic plots for visualization of ergodic partition and invariant sets
• Mercer’s Theorem, Feature Maps, and Smoothing
• Hessian eigenmaps: Locally linear embedding techniques for high-dimensional data
• Theory of Reproducing Kernels