Multi-scale geometric methods for data sets. II: Geometric multi-resolution analysis (Q413654): Difference between revisions
From MaRDI portal
Created a new Item |
ReferenceBot (talk | contribs) Changed an Item |
||
(8 intermediate revisions by 6 users not shown) | |||
Property / author | |||
Property / author: Mauro Maggioni / rank | |||
Property / author | |||
Property / author: Mauro Maggioni / rank | |||
Normal rank | |||
Property / review text | |||
The concept of a {\textit{geometric multiresolution analysis}} (GMRA) is introduced as a means of analyzing data sets modeled as lying on a potentially low \(d\)-dimensional set inside \(\mathbb{R}^D\), \(D\) large. The construction consists of three main steps: (1) a multi-scale geometric tree decomposition of a given metric space \(\mathcal{M}\) into subsets \(\{C_{j,k}\}_{k\in \mathcal{K}_j,\, j\in\mathbb{Z}}\) called {\textit{dyadic cells}}; (2) a \(d\)-dimensional affine approximation of each \(C_{j,k}\), giving rise to a sequence \(\{\mathcal{M}_j\}_{j\in\mathbb{Z}}\) of piecewise linear sets ({\textit{scaling-like spaces}}); and (3) low-dimensional affine encodings of the differences between \(\mathcal{M}_{j+1}\) and \(\mathcal{M}_j\) (wavelet-like projections). Here, \((\mathcal{M},\mu,\rho)\) is a Borel probability space with \(\mathcal{M}\subset\mathbb{R}^D\). The sequence \(\{C_{j,k}\}\) is analogous to a dyadic cube decomposition. The term geometric wavelets applies because the definitions build on classical multi-scale geometric measure theory. For \(c_{j,k}=\frac{1}{\mu(C_{j,k})}\int_{C_{j,k}} x\, d\mu\) the affine space \(\mathbb{V}_{j,k}=c_{j,k}+V_{j,k}\) with \(V_{j,k}\) spanned by the columns of the matrix \(\Phi_{j,k}\) in the singular value decomposition \(\text{cov}_{j,k}= \Phi\Sigma\Phi^\ast\) of the covariance \(\text{cov}_{j,k}=\mathbb{E}_\mu[(x-c_{j,k})(x-c_{j,k})^\ast|\, x\in C_{j,k}]\) can be thought of as an approximate tangent space to \(\mathcal{M}\) at \(c_{j,k}\) at scale \(j\). Let \(P_{j,k}=\Phi_{j,k}\Phi_{j,k}^\ast\) be the projection matrix onto \(V_{j,k}\) and define the geometric wavelet subspace \(W_{j+1,k}=(I-P_{j,k}) V_{j+1,k}\) and \(\Psi_{j+1,k}\) be the orthogonal matrix of this projection. For \(x\in C_{j,k}\) one sets \(x_j=P_{\mathcal{M}_j}(x)=P_{j,k}(x-c_{j,k})+c_{j,k}\) and \(Q_{\mathcal{M}_{j+1}}(x)=x_{j+1}-x_j\). Then \(P_{\mathcal{M}_{j+1}}=P_{\mathcal{M}_j}+Q_{\mathcal{M}_{j+1}}\) and one says that together the sequence of affine operators forms a GMRA. Further details are worked out in the case of manifolds and {\textit{point cloud}} data and generic algorithms are provided for numerical construction of GMRAs and fast geometric wavelet transforms and their inverses. A series of numerical examples, both synthetic and data driven, is provided. After this, a refined construction of {\textit{orthogonal}} GMRAs is provided which removes redundancies across scales. Again, pseudo-code is provided for this construction. A series of useful variations leading to refined numerical analysis is also provided together with computation cost guidelines. | |||
Property / review text: The concept of a {\textit{geometric multiresolution analysis}} (GMRA) is introduced as a means of analyzing data sets modeled as lying on a potentially low \(d\)-dimensional set inside \(\mathbb{R}^D\), \(D\) large. The construction consists of three main steps: (1) a multi-scale geometric tree decomposition of a given metric space \(\mathcal{M}\) into subsets \(\{C_{j,k}\}_{k\in \mathcal{K}_j,\, j\in\mathbb{Z}}\) called {\textit{dyadic cells}}; (2) a \(d\)-dimensional affine approximation of each \(C_{j,k}\), giving rise to a sequence \(\{\mathcal{M}_j\}_{j\in\mathbb{Z}}\) of piecewise linear sets ({\textit{scaling-like spaces}}); and (3) low-dimensional affine encodings of the differences between \(\mathcal{M}_{j+1}\) and \(\mathcal{M}_j\) (wavelet-like projections). Here, \((\mathcal{M},\mu,\rho)\) is a Borel probability space with \(\mathcal{M}\subset\mathbb{R}^D\). The sequence \(\{C_{j,k}\}\) is analogous to a dyadic cube decomposition. The term geometric wavelets applies because the definitions build on classical multi-scale geometric measure theory. For \(c_{j,k}=\frac{1}{\mu(C_{j,k})}\int_{C_{j,k}} x\, d\mu\) the affine space \(\mathbb{V}_{j,k}=c_{j,k}+V_{j,k}\) with \(V_{j,k}\) spanned by the columns of the matrix \(\Phi_{j,k}\) in the singular value decomposition \(\text{cov}_{j,k}= \Phi\Sigma\Phi^\ast\) of the covariance \(\text{cov}_{j,k}=\mathbb{E}_\mu[(x-c_{j,k})(x-c_{j,k})^\ast|\, x\in C_{j,k}]\) can be thought of as an approximate tangent space to \(\mathcal{M}\) at \(c_{j,k}\) at scale \(j\). Let \(P_{j,k}=\Phi_{j,k}\Phi_{j,k}^\ast\) be the projection matrix onto \(V_{j,k}\) and define the geometric wavelet subspace \(W_{j+1,k}=(I-P_{j,k}) V_{j+1,k}\) and \(\Psi_{j+1,k}\) be the orthogonal matrix of this projection. For \(x\in C_{j,k}\) one sets \(x_j=P_{\mathcal{M}_j}(x)=P_{j,k}(x-c_{j,k})+c_{j,k}\) and \(Q_{\mathcal{M}_{j+1}}(x)=x_{j+1}-x_j\). Then \(P_{\mathcal{M}_{j+1}}=P_{\mathcal{M}_j}+Q_{\mathcal{M}_{j+1}}\) and one says that together the sequence of affine operators forms a GMRA. Further details are worked out in the case of manifolds and {\textit{point cloud}} data and generic algorithms are provided for numerical construction of GMRAs and fast geometric wavelet transforms and their inverses. A series of numerical examples, both synthetic and data driven, is provided. After this, a refined construction of {\textit{orthogonal}} GMRAs is provided which removes redundancies across scales. Again, pseudo-code is provided for this construction. A series of useful variations leading to refined numerical analysis is also provided together with computation cost guidelines. / rank | |||
Normal rank | |||
Property / Mathematics Subject Classification ID | |||
Property / Mathematics Subject Classification ID: 42C99 / rank | |||
Normal rank | |||
Property / Mathematics Subject Classification ID | |||
Property / Mathematics Subject Classification ID: 42C40 / rank | |||
Normal rank | |||
Property / Mathematics Subject Classification ID | |||
Property / Mathematics Subject Classification ID: 68T05 / rank | |||
Normal rank | |||
Property / zbMATH DE Number | |||
Property / zbMATH DE Number: 6031284 / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
multi-scale analysis | |||
Property / zbMATH Keywords: multi-scale analysis / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
multiresolution analysis | |||
Property / zbMATH Keywords: multiresolution analysis / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
geometric wavelets | |||
Property / zbMATH Keywords: geometric wavelets / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
high-dimensional data set | |||
Property / zbMATH Keywords: high-dimensional data set / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
frame | |||
Property / zbMATH Keywords: frame / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
dictionary learning | |||
Property / zbMATH Keywords: dictionary learning / rank | |||
Normal rank | |||
Property / reviewed by | |||
Property / reviewed by: Joseph D. Lakey / rank | |||
Normal rank | |||
Property / describes a project that uses | |||
Property / describes a project that uses: PDCO / rank | |||
Normal rank | |||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank | |||
Property / OpenAlex ID | |||
Property / OpenAlex ID: W2964012263 / rank | |||
Normal rank | |||
Property / arXiv ID | |||
Property / arXiv ID: 1105.4924 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Multiscale Representations for Manifold-Valued Data / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Geometric diffusions as a tool for harmonic analysis and structure definition of data: Diffusion maps / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Geometric diffusions as a tool for harmonic analysis and structure definition of data: Multiscale methods / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Multiscale geometric methods for data sets. I: Multiscale SVD, noise and curvature. / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Intrinsic dimension estimation of data: An approach based on Grassberger-Procaccia's algorithm / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Hessian eigenmaps: Locally linear embedding techniques for high-dimensional data / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Principal Manifolds and Nonlinear Dimensionality Reduction via Tangent Space Alignment / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Manifold parametrizations by eigenfunctions of the Laplacian and heat kernels / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Diffusion wavelets / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Diffusion wavelet packets / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q3174155 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q3096167 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q5486952 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4359338 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Atomic Decomposition by Basis Pursuit / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Ten Lectures on Wavelets / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: An introduction to frames and Riesz bases / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Image decomposition via the combination of sparse representations and a variational approach / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4656813 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q2896018 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4864293 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Rectifiable sets and the traveling salesman problem / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4278678 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Fast computation in adaptive tree approximation / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q3546378 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: A theory for multiresolution signal decomposition: the wavelet representation / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Multiresolution Approximations and Wavelet Orthonormal Bases of L 2 (R) / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4705314 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q3995454 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: A T(b) theorem with remarks on analytic capacity and the Cauchy integral / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Diffusion maps / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Fast high-dimensional approximation with sparse occupancy trees / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Curvature Measures / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Random projections of smooth manifolds / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Finding the homology of submanifolds with high confidence from random samples / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: The traveling salesman problem and harmonic analysis / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Uniform rectifiability and quasiminimizing sets of arbitrary codimension / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q3359644 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Asymptotic regularity of subdivisions of Euclidean domains by iterated PCA and iterated 2-means / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4849992 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: A Randomized Algorithm for Principal Component Analysis / rank | |||
Normal rank | |||
links / mardi / name | links / mardi / name | ||
Latest revision as of 03:27, 5 July 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Multi-scale geometric methods for data sets. II: Geometric multi-resolution analysis |
scientific article |
Statements
Multi-scale geometric methods for data sets. II: Geometric multi-resolution analysis (English)
0 references
7 May 2012
0 references
The concept of a {\textit{geometric multiresolution analysis}} (GMRA) is introduced as a means of analyzing data sets modeled as lying on a potentially low \(d\)-dimensional set inside \(\mathbb{R}^D\), \(D\) large. The construction consists of three main steps: (1) a multi-scale geometric tree decomposition of a given metric space \(\mathcal{M}\) into subsets \(\{C_{j,k}\}_{k\in \mathcal{K}_j,\, j\in\mathbb{Z}}\) called {\textit{dyadic cells}}; (2) a \(d\)-dimensional affine approximation of each \(C_{j,k}\), giving rise to a sequence \(\{\mathcal{M}_j\}_{j\in\mathbb{Z}}\) of piecewise linear sets ({\textit{scaling-like spaces}}); and (3) low-dimensional affine encodings of the differences between \(\mathcal{M}_{j+1}\) and \(\mathcal{M}_j\) (wavelet-like projections). Here, \((\mathcal{M},\mu,\rho)\) is a Borel probability space with \(\mathcal{M}\subset\mathbb{R}^D\). The sequence \(\{C_{j,k}\}\) is analogous to a dyadic cube decomposition. The term geometric wavelets applies because the definitions build on classical multi-scale geometric measure theory. For \(c_{j,k}=\frac{1}{\mu(C_{j,k})}\int_{C_{j,k}} x\, d\mu\) the affine space \(\mathbb{V}_{j,k}=c_{j,k}+V_{j,k}\) with \(V_{j,k}\) spanned by the columns of the matrix \(\Phi_{j,k}\) in the singular value decomposition \(\text{cov}_{j,k}= \Phi\Sigma\Phi^\ast\) of the covariance \(\text{cov}_{j,k}=\mathbb{E}_\mu[(x-c_{j,k})(x-c_{j,k})^\ast|\, x\in C_{j,k}]\) can be thought of as an approximate tangent space to \(\mathcal{M}\) at \(c_{j,k}\) at scale \(j\). Let \(P_{j,k}=\Phi_{j,k}\Phi_{j,k}^\ast\) be the projection matrix onto \(V_{j,k}\) and define the geometric wavelet subspace \(W_{j+1,k}=(I-P_{j,k}) V_{j+1,k}\) and \(\Psi_{j+1,k}\) be the orthogonal matrix of this projection. For \(x\in C_{j,k}\) one sets \(x_j=P_{\mathcal{M}_j}(x)=P_{j,k}(x-c_{j,k})+c_{j,k}\) and \(Q_{\mathcal{M}_{j+1}}(x)=x_{j+1}-x_j\). Then \(P_{\mathcal{M}_{j+1}}=P_{\mathcal{M}_j}+Q_{\mathcal{M}_{j+1}}\) and one says that together the sequence of affine operators forms a GMRA. Further details are worked out in the case of manifolds and {\textit{point cloud}} data and generic algorithms are provided for numerical construction of GMRAs and fast geometric wavelet transforms and their inverses. A series of numerical examples, both synthetic and data driven, is provided. After this, a refined construction of {\textit{orthogonal}} GMRAs is provided which removes redundancies across scales. Again, pseudo-code is provided for this construction. A series of useful variations leading to refined numerical analysis is also provided together with computation cost guidelines.
0 references
multi-scale analysis
0 references
multiresolution analysis
0 references
geometric wavelets
0 references
high-dimensional data set
0 references
frame
0 references
dictionary learning
0 references
0 references
0 references
0 references
0 references
0 references