Linear/ridge expansions: enhancing linear approximations by ridge functions

DOI10.1007/S10444-022-09936-4zbMATH Open1490.65022arXiv2107.04345OpenAlexW3180295362WikidataQ114227747 ScholiaQ114227747MaRDI QIDQ2124754FDOQ2124754

Constantin Greif, Philipp Junk, Karsten Urban

Publication date: 11 April 2022

Published in: Advances in Computational Mathematics (Search for Journal in Brave)

Abstract: We consider approximations formed by the sum of a linear combination of given functions enhanced by ridge functions -- a Linear/Ridge expansion. For an explicitly or implicitly given function, we reformulate finding a best Linear/Ridge expansion in terms of an optimization problem. We introduce a particle grid algorithm for its solution. Several numerical results underline the flexibility, robustness and efficiency of the algorithm. One particular source of motivation is model reduction of parameterized transport or wave equations. We show that the particle grid algorithm is able to produce a Linear/Ridge expansion as an efficient nonlinear model reduction.

Full work available at URL: https://arxiv.org/abs/2107.04345

Recommendations

• Data-driven polynomial ridge approximation using variable projection
• On the representation by bivariate ridge functions
• Ridge approximations and the Kolmogorov-Nikol'skij problem.
• Publication:4938096
• Approximation of generalized ridge functions in high dimensions

zbMATH Keywords

linear approximationnonlinear model reductionridge functionsparticle grid algorithm

Mathematics Subject Classification ID

Algorithms for approximation of functions (65D15) Approximation by arbitrary nonlinear expressions; widths and entropy (41A46)

Cites Work

• Reduced Basis Methods for Partial Differential Equations
• Title not available (Why is that?)
• Some observations on Babuška and Brezzi theories
• Convergence Rates for Greedy Algorithms in Reduced Basis Methods
• Title not available (Why is that?)
• \textit{A priori} convergence of the greedy algorithm for the parametrized reduced basis method
• Approximation variationnelle des problèmes aux limites
• Ridge Functions
• Identifying linear combinations of ridge functions
• Certified Reduced Basis Methods for Parametrized Partial Differential Equations
• Nonlinear reduced basis approximation of parameterized evolution equations via the method of freezing
• Kolmogorov widths and low-rank approximations of parametric elliptic PDEs
• Title not available (Why is that?)
• Data driven approximation of parametrized PDEs by reduced basis and neural networks
• A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized PDEs
• Nonlinear methods for model reduction
• On some aspects of approximation of ridge functions
• Decay of the Kolmogorov \(N\)-width for wave problems
• A theoretical analysis of deep neural networks and parametric PDEs
• Transformed Snapshot Interpolation with High Resolution Transforms
• A near-stationary subspace for ridge approximation
• Nonlinear model reduction on metric spaces. Application to one-dimensional conservative PDEs in Wasserstein spaces
• Projection-based model reduction with dynamically transformed modes

Cited In (3)

• The reduced basis method in space and time: challenges, limits and perspectives
• Identifying linear combinations of ridge functions
• On the representability of a continuous multivariate function by sums of ridge functions

Uses Software

• redbKIT

This page was built for publication: Linear/ridge expansions: enhancing linear approximations by ridge functions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2124754)