Interpretable approximation of high-dimensional data
From MaRDI portal
Approximate quadratures (41A55) Multidimensional problems (41A63) Fourier series and coefficients in several variables (42B05) Analysis of variance and covariance (ANOVA) (62J10) Algorithms for approximation of functions (65D15) Numerical methods for trigonometric approximation and interpolation (65T40) Numerical methods for discrete and fast Fourier transforms (65T50)
Abstract: In this paper we apply the previously introduced approximation method based on the ANOVA (analysis of variance) decomposition and Grouped Transformations to synthetic and real data. The advantage of this method is the interpretability of the approximation, i.e., the ability to rank the importance of the attribute interactions or the variable couplings. Moreover, we are able to generate an attribute ranking to identify unimportant variables and reduce the dimensionality of the problem. We compare the method to other approaches on publicly available benchmark datasets.
Recommendations
- Grouped transformations and regularization in high-dimensional explainable ANOVA approximation
- Approximation of high-dimensional periodic functions with Fourier-based methods
- Learning multivariate functions with low-dimensional structures using polynomial bases
- Interpretable dimension reduction
- Anchor points matter in ANOVA decomposition
Cites work
- 10.1162/153244303322753616
- \(L_2\)-norm sampling discretization and recovery of functions from RKHS with finite trace
- A near-stationary subspace for ridge approximation
- Application of quasi-Monte Carlo methods to elliptic PDEs with random diffusion coefficients: a survey of analysis and implementation
- Approximation of functions of few variables in high dimensions
- Approximation of high-dimensional periodic functions with Fourier-based methods
- Circulant embedding with QMC: analysis for elliptic PDE with lognormal coefficients
- Data classification. Algorithms and applications
- Effective dimension of some weighted pre-Sobolev spaces with dominating mixed partial derivatives
- Erratum: Active Subspace Methods in Theory and Practice: Applications to Kriging Surfaces
- Estimating Mean Dimensionality of Analysis of Variance Decompositions
- Experiments: planning, analysis, and optimization
- Fast high-dimensional approximation with sparse occupancy trees
- General foundations of high-dimensional model representations
- Global sensitivity analysis: The primer
- Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates
- scientific article; zbMATH DE number 2034576 (Why is no real title available?)
- Learning functions of few arbitrary linear parameters in high dimensions
- Learning multivariate functions with low-dimensional structures using polynomial bases
- LSQR: An Algorithm for Sparse Linear Equations and Sparse Least Squares
- Multivariate regression and machine learning with sums of separable functions
- Numerical Fourier analysis
- On decompositions of multivariate functions
- On sensitivity estimation for nonlinear mathematical models
- Pattern recognition and machine learning.
- Quasi-Monte Carlo finite element methods for a class of elliptic partial differential equations with random coefficients
- Quasi-Monte Carlo finite element methods for elliptic PDEs with lognormal random coefficients
- Scikit-learn: machine learning in Python
- Sparse grid quadrature in high dimensions with applications in finance and insurance
- Towards a unified analysis of random Fourier features
- User-friendly tail bounds for sums of random matrices
- Using NFFT 3 -- a software library for various nonequispaced fast Fourier transforms
Cited in
(11)- Integrative exploration of large high-dimensional datasets
- High-dimensional approximate r-nets
- Grouped transformations and regularization in high-dimensional explainable ANOVA approximation
- HARFE: hard-ridge random feature expansion
- Fast hyperbolic wavelet regression meets ANOVA
- On the approximability and curse of dimensionality of certain classes of high-dimensional functions
- Variable transformations in combination with wavelets and ANOVA for high-dimensional approximation
- ANOVA approximation with mixed tensor product basis on scattered points
- MARS via lasso
- Interpretable linear dimensionality reduction based on bias-variance analysis
- Fast and interpretable support vector classification based on the truncated ANOVA decomposition
This page was built for publication: Interpretable approximation of high-dimensional data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5018903)
