Iterated diffusion maps for feature identification
From MaRDI portal
Abstract: Recently, the theory of diffusion maps was extended to a large class of local kernels with exponential decay which were shown to represent various Riemannian geometries on a data set sampled from a manifold embedded in Euclidean space. Moreover, local kernels were used to represent a diffeomorphism, H, between a data set and a feature of interest using an anisotropic kernel function, defined by a covariance matrix based on the local derivatives, DH. In this paper, we generalize the theory of local kernels to represent degenerate mappings where the intrinsic dimension of the data set is higher than the intrinsic dimension of the feature space. First, we present a rigorous method with asymptotic error bounds for estimating DH from the training data set and feature values. We then derive scaling laws for the singular values of the local linear structure of the data, which allows the identification the tangent space and improved estimation of the intrinsic dimension of the manifold and the bandwidth parameter of the diffusion maps algorithm. Using these numerical tools, our approach to feature identification is to iterate the diffusion map with appropriately chosen local kernels that emphasize the features of interest. We interpret the iterated diffusion map (IDM) as a discrete approximation to an intrinsic geometric flow which smoothly changes the geometry of the data space to emphasize the feature of interest. When the data lies on a product manifold of the feature manifold with an irrelevant manifold, we show that the IDM converges to the quotient manifold which is isometric to the feature manifold, thereby eliminating the irrelevant dimensions. We will also demonstrate empirically that if we apply the IDM to features that are not a quotient of the data space, the algorithm identifies an intrinsically lower-dimensional set embedding of the data which better represents the features.
Recommendations
Cites work
- Diffusion maps
- Embedding Riemannian manifolds by their heat kernel
- Embeddings of Riemannian manifolds with heat kernels and eigenfunctions
- From graph to manifold Laplacian: the convergence rate
- Graph Laplacian Tomography From Unknown Random Projections
- Laplacian Eigenmaps for Dimensionality Reduction and Data Representation
- Local kernels and the geometric structure of data
- Non-linear independent component analysis with diffusion maps
- The Laplacian on a Riemannian manifold. An introduction to analysis on manifolds
- Variable bandwidth diffusion kernels
- Vector diffusion maps and the connection Laplacian
Cited in
(12)- Spectral methods for solving elliptic PDEs on unknown manifolds
- Diffusion maps: using the semigroup property for parameter tuning
- Reproducing kernel Hilbert space compactification of unitary evolution groups
- Diffusion maps tailored to arbitrary non-degenerate Itô processes
- Approximating solutions of linear elliptic PDE's on a smooth manifold using local kernel
- Supervised dimension reduction via local gradient elongation
- Fractional diffusion maps
- Kernel-based methods for solving time-dependent advection-diffusion equations on manifolds
- Ghost Point Diffusion Maps for Solving Elliptic PDEs on Manifolds with Classical Boundary Conditions
- Augmentation invariant manifold learning
- Applications of topological graph theory to 2-manifold learning
- Manifold learning for organizing unstructured sets of process observations
This page was built for publication: Iterated diffusion maps for feature identification
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1748254)
